2014The drivers and impediments for cross-border e-commerce in the EU

The drivers and impediments for cross-border e-commerce in the

EU

Estrella Gomez-Herrera ?,Bertin Martens 1,Geomina Turlea

Institute for Prospective Technological Studies,Joint Research Centre,European Commission,C/Inca Garcilaso 3,E-41092Seville,Spain

a r t i c l e i n f o Article history:

Received 6May 2013

Received in revised form 30April 2014Accepted 16May 2014

Available online 2June 2014JEL classi?cation:F15O52

Keywords:E-commerce

Gravity equation European Union

a b s t r a c t

The rise of the internet is often associated with the ‘‘death of distance’’or at least the decreasing relevance of geographical distance in the supply of information.We investigate whether distance still matters for online trade in physical goods.We use data from an online consumer survey panel on online cross-border trade in goods in a linguistically frag-mented EU market.The analysis con?rms that distance-related trade costs are greatly reduced compared to of?ine trade in the same goods.However,language-related trade costs increase.Moreover,online trade introduces new sources of trade costs such as parcel delivery and online payments systems.On balance,there are no indications that online trade is less biased in favour of home market products than of?ine trade.We examine options available to policy makers to boost cross-border e-commerce in the EU Digital Sin-gle Market.A 1%increase in the use of ef?cient and ?exible cross-border payment systems could increase cross-border e-commerce by as much as 7%.We also show that online trade gives a comparative advantage to English-language exporting countries.

ó2014Elsevier B.V.All rights reserved.

1.Introduction

This paper empirically investigates the impact of online e-commerce on cross-border trade patterns.The rise of the internet and,more generally,digital communications technology,has led many observers to announce the ‘‘death of distance’’(Cairncross,1997).In this view,it does not matter anymore where information is located since it is only a mouse click away and information costs are no longer related to physical distance.For traditional of?ine trade in physical goods however,the evidence actually points to an increase in distance costs (Disdier and Head,2008).Trade is based on a combination of information and physical shipping of goods.The question is whether

shifting trade from of?ine to online platforms makes a suf?ciently large dent in information costs to change total trade costs and therefore the pattern of trade in goods.Blum and Goldfarb (2006)show that even for pure information products,distance still plays a signi?cant role.They attribute this to cultural differences that increase with physical distance.Apart from information cost effects,there may be secondary effects that affect trade patterns.Online trade opens up a potentially much larger geograph-ical catchment area,both for suppliers and consumers,with an increase in variety of available products and in price competition.Both factors would favour a relative shift away from of?ine and towards online trade.However,new sources of information trade costs may arise online that may slow down or even reverse this trend.New infor-mation costs may be attributable to linguistic,cultural and institutional differences and the trade costs related to the operations of e-commerce infrastructure.

https://www.wendangku.net/doc/7d3284606.html,/10.1016/https://www.wendangku.net/doc/7d3284606.html,ecopol.2014.05.0020167-6245/ó2014Elsevier B.V.All rights reserved.

?

Corresponding author.Tel.:+34954488456.

E-mail addresses:estrellagh@https://www.wendangku.net/doc/7d3284606.html, (E.Gomez-Herrera),Bertin.MARTENS@ec.europa.eu (B.Martens),Geomina@yahoo.fr (G.Turlea).1

Tel.:+34954488366.

There is so far very little research on cross-border e-commerce in physical goods and the interaction between physical trade costs and information costs.The absence of of?cial statistics on cross-border e-commerce in goods has so far limited empirical work on this subject to a few cases of privileged access to proprietary datasets.We go beyond previous research that focused on the reduction in distance costs in online trade,either for domestic e-commerce (Horta?su et al.,2009)or for international trade in infor-mation services(Blum and Goldfarb,2006).We follow Lendle et al.(2012)who study gravity effects in online cross-border trade in goods,using a proprietary eBay data-base.This article differs from previous literature in two main respects.First,we use a general online consumer sur-vey that is not linked to a speci?c e-commerce platform. Second,we focus on intra-EU e-commerce to disentangle the main barriers that still affect the Digital Single Market. Moreover,we combine distance related trade costs with other sources of trade costs to estimate home bias and the border effect in online trade,building on McCallum (1995),Wolf(2000),Pacchioli(2011).

We apply this framework to a unique dataset of cross-border e-commerce in goods obtained from an online con-sumer survey(Civic Consulting,2011)in a linguistically fragmented EU market to explore policy options to boost the EU Digital Single Market.According to the European Commission(2012),ten years after the adoption of the EU E-Commerce Directive,e-commerce is still limited to less than4%of total European cross-border e-commerce trade and considers that this is far below its full potential. The Commission’s Digital Agenda for Europe aims to get 50%of all European citizens to buy online and20%to engage in online cross-border transactions by2015.The question is whether the potential for cross-border transac-tions is higher in e-commerce than in of?ine trade.

We investigate three potential sources of changes in online trade costs,compared to of?ine trade.First,the shift from ordinary of?ine trade to internet-enabled online trade may reduce the importance of geographical dis-tance-related trade costs.While distance may no longer matter for information and purely digital products and services(Blum and Goldfarb,2006),goods still need to be physically transported,and sometimes cross borders between different regulatory regimes,to reach the buyer. Consequently,only part of the total trade costs is affected by the shift from analogue to digital information technology.Second,we assess the role of cultural and insti-tutional factors,such as language and the quality of legal institutions,as determinants of online trade patterns.As distance-related trade costs diminish,the relative impor-tance of other sources of costs may increase.Third,online trading platforms for physical goods require speci?c infra-structure,such as?exible online payment systems and cost-ef?cient parcel delivery systems.We gauge their con-tribution to explaining online trade patterns.Finally,we combine all these sources of online trade costs and look at the net effect of positive and negative contributions,as measured by the degree of home bias or the‘‘natural’’pref-erence for home market products.

The analytical tool that we use for this purpose is the gravity model of cross-border international trade,the standard workhorse for explaining international trade ?ows in the of?ine economy(Anderson and Van Wincoop,2003).This model is rooted in the Newtonian idea that many of the observed patterns of international trade?ows can be explained by the economic size of the trading partners and their physical distance.‘‘Distance’’can be more broadly interpreted as a catch-all variable and proxy for various sources of international trade costs that affect the relative price of domestic and imported goods.This may include physical transport costs,the cost associated with import tariffs and regulatory barriers, and risks related to poor contract enforcement between different jurisdictions.In a traditional bricks and mortar economy,information retrieval is costly and requires physical transport,either to bring information to potential customers or vice versa.Here,we try to separate the infor-mation cost from the physical transport cost dimension.

We?nd that that the importance of geographical distance-related trade costs is indeed greatly reduced in online trade,compared to of?ine trade.On the other hand, socio-cultural variables such as language increase in importance and counterbalance the declining cost of distance.Moreover,other sources of trade costs gain in prominence for online transactions,in particular payments and parcel delivery systems.Overall,there are no indica-tions that home bias is less signi?cant online than of?ine, if we compare our online results with others in the of?ine trade literature.This may be due to the fact that consumers (in a business-to-consumer(B2C)online trade setting)are more sensitive to these new sources of trade costs than businesses(in a business-to-business(B2B)of?ine trade setting)dealing with each other in more established of?ine relationships.We are cautious however in interpreting these?ndings because the supply chain of intermediaries involved in online trade clearly differs from those involved in of?ine trade.

The article is structured as follows.The next section presents a brief literature overview of the existing litera-ture on international online trade and the use of gravity models in international trade,including the role and inter-pretation of the distance variable in these models.Section3 discusses the gravity model that we apply.Section4 explains the data sources for the model.The construction of the bilateral online trade matrix is explained in Appen-dix A.Section5presents the estimation results.Finally, Section6summarizes and presents some policy-related conclusions.

2.Literature review

Distance is a key variable in international trade models. It is basically a catch-all term that proxies various sources of cross-border trade costs that affect the relative price of domestic and imported goods.This may include transport costs,import tariffs,differences in technical standards and regulatory regimes between countries that induce additional trade costs,and risks related to poor institu-tional quality and weak contract enforcement across borders.The role of distance in international trade remains a dif?cult issue.Despite the decline in international transport costs and especially communication costs,the

84 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

importance of distance does not appear to decline in gravity estimates over longer time periods.Disdier and Head(2008)perform a meta-analysis on1467estimates of the distance elasticity in gravity models.They conclude that the importance of distance decreased in international trade between1870and1950but since then it has been rising again.Berthelon and Freund(2008)show that this increase is not related to changes in the composition of trade.The importance of distance has increased for about 25%of all goods,mostly for homogenous goods sold on bulk exchange https://www.wendangku.net/doc/7d3284606.html,rmation on homogenous goods is easier to convey,mainly thanks to the fall in information costs.That results in a relative decline in the importance of information costs and a relative increase in the importance of distance-related transport costs.By contrast,differenti-ated products require more information and are thus rela-tively less sensitive to distance-related transport costs. Despite the fall in information costs,the ratio of distance to non-distance related trade costs for differentiated goods seems to have remained fairly constant.In contrast,Rauch (1999)argues that the fall in communication costs had a greater impact on differentiated goods than on homoge-nous goods.

The impact of distance on trade may be subject to dis-continuities associated with the organisation of human societies.Regional differences due to borders,cultural, institutional and other types of geographically con?ned variables may affect trade costs.The so-called‘‘border effect’’aggregates the combined impact of all these vari-ables.The higher the trade costs related to crossing a bor-der,the less outward-oriented and the more home-based trade patterns will be.McCallum(1995)applies the gravity model to trade between Canadian provinces and US states. He?nds that although Canadian provinces are often closer to neighbouring US states than to neighbouring provinces, the US–Canada border still constitutes a signi?cant source of trade costs and thus a barrier to trade.Wolf(2000)uses gravity to estimate home bias as an alternative measure of border costs in trade between US states.As language,cul-ture,regulatory regimes and technical standards are pretty much similar across US states,at least much more so than between the US and other countries,one would expect the border effect or home bias to disappear in trade?ows between US states.His research shows that home bias is indeed substantially lower in intra-US trade than in intra-OECD trade but remains a signi?cant trade barrier. In the absence of regulatory differences,this may simply re?ect a‘‘natural’’degree of consumer preference for local suppliers.Closeness of buyers and sellers may enhance the perception of trust,veri?cation of product quality and eas-ier settlement of disputes.Coughlin and Novy(2009) extend Wolf’s research to cover both domestic trade between US states and international trade between US states and foreign trade partners.Somewhat surprisingly, they?nd that the domestic border between US states con-stitutes a larger trade barrier than crossing the interna-tional US border.Frankel and Wei(1995)apply the gravity model to trade between EU countries and conclude that trade costs at intra-EU borders are signi?cant despite the fact that all these countries belong to a customs union. Pacchioli(2011)compares home bias in the US and in the EU internal market,as a proxy measure of the success of the EU’s drive to complete the Single Market.She uses data on trade?ows between EU Member States and between states in the US.She?nds a higher degree of home bias in the EU than in the US and concludes that there is still some way to go for the EU Single Market.

All the above work was conducted on traditional of?ine trade data.Empirical work on online trade?ows has been very limited so far,mainly because of the absence of of?cial statistics on online cross-border trade.Data are generated mainly by private companies involved in online https://www.wendangku.net/doc/7d3284606.html,mercial interests stop them from publishing these data.A few empirical research articles have tried to circumvent this data gap and examine cross-border online operations from various indirect angles.Freund and Weinhold(2000)examine how internet penetration in countries affects their ordinary of?ine trade patterns.Blum and Goldfarb(2006)try to explain interna-tional internet click stream patterns using a gravity model. Their research focuses on online digital information prod-ucts that can be transported across the internet at zero trade cost–anything but physical goods that need physical transport to reach the consumer.Still,they?nd that geographical distance plays a relevant role in these international purely information trade patterns.This?nd-ing is particularly true for digital products that depend on what they call‘‘taste’’.Distance decreases the likelihood of a shared cultural context.For less taste-and culture-dependent products distance has no statistically signi?cant effect in their?ndings.Horta?su et al.(2009)are the?rst to look at actual online transactions in physical goods.They take a sample of intra-US trade observations from eBay and cross-border trade from MercadoLibre to examine the importance of distance in these transactions.They con-clude that distance still has an impact on trade,though less so in online than in of?ine transactions.Lendle et al.(2012) use eBay data on cross-border transactions between62 countries for the period2004–2007to estimate a gravity model of online trade with several explanatory variables, including distance,transport costs,common language,bor-der,legal regime or colonial background and quality of governance.They?nd that nearly all these factors generate less trade costs on eBay than in of?ine trade,except for language and shipping costs.However,they do not investi-gate the combined effect of increases and decreases in trade costs.

3.The gravity model

In order to make sense out of the bilateral trade data generated by the consumer survey,we need an explana-tory model for these trade?ows.In line with previous research on online trade(Blum and Goldfarb,2006; Horta?su et al.,2009;Lendle et al.,2012)we apply the well-known gravity model,the workhorse of international trade modelling.Over the last decade the gravity model has undergone substantial improvement and modi?ca-tions,especially after Anderson and Van Wincoop(2003) provided more solid theoretical underpinnings.Since then it has become widely used in trade economics and beyond.

E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–9685

The gravity model explains the value of bilateral trade (T ij)between two countries i and j as a function of the product of the size of the two economies(proxied by GDP)and the distance(D ij)between them.This approach is essentially the equivalent of Newton’s gravity theory from physics:

T ij?aeGDP i x GDP jT=D ije1TThe advantage of putting this model in log–log format is that the coef?cients become elasticities.The value of b for instance is the percentage change in cross-border trade T ij induced by a one-percent change in GDP i.

ln T ij?b0tb1ln GDP itb2ln GDP jtb3ln D ijte ije2T

A key issue with this speci?cation of the gravity model is that trade between countries i and j is not only a function of speci?c bilateral factors but is also affected by the presence and trade costs of all other countries,or by multilateral effects.McCallum(1995)uses‘‘remoteness’’as a measure of multilateral resistance,i.e.the weighted average distance between a country and all its trading partners.Anderson and Van Wincoop(2003)propose the inclusion of complete set of multilateral resistance terms derived from all the variables in their model.Baier and Bergstrand(2009)introduce multilateral resistance terms for each variable,calculated as the GDP-weighted average of the values for the relevant variables across all other trading partners.Alternatively,Feenstra(2002)proposes the introduction of importer and exporter country?xed effects.Country dummies will capture both multilateral resistance and any other country speci?c characteristic that may in?uence trade.The main drawback of this approach is that in a cross-section analysis the GDP coef?-cients are no longer observed.We adopt the Feenstra (2002)approach and introduce country speci?c?xed effects.Hence,the?nal speci?cation of our variable is as follows:

ln T ij?b0tb1ln D ijtg itg jte ije3T

where g i and g j are a set of dummies for the importer and the exporter correspondingly.

Apart from the speci?cation,Santos Silva and Tenreyro (2006)question the estimation methods for gravity mod-els.They argue that log-linearised gravity equations are potentially subject to biased estimation because of heter-oskedasticity in the error term and zero values for some observations.Heteroskedasticity tests show that the errors are not always normally distributed.Moreover,the loga-rithmic version of the gravity equation forces us to drop zeros and not available trade observations from the esti-mation because the logarithm of zero is not de?ned.This discards potentially useful information.Santos Silva and Tenreyro(2006)demonstrate that Poisson pseudo maxi-mum likelihood(PPML)is a less biased and more ef?cient estimator that avoids the problems of zero observations and heteroskedasticity.

However,PPML does not provide an explanation for zero trade observations.Two-step estimators have been introduced in the recent literature to overcome this.The decision to trade or not and the decision on how much to trade are not completely independent decisions.Two-step models allow for some positive correlation between the error term in the selection and regression equations to bet-ter re?ect the real decision process.The Helpman,Melitz and Rubinstein(2008)(HMR henceforth)two-step approach has a particular importance.The?rst step con-sists of the estimation of the probability of exporting using a Probit model.Variables affecting the probability of exporting but not the size of exports(exclusion variables) are included.HMR chooses variables related to trade barri-ers that affect?xed trade costs but do not affect variable (per-unit)trade https://www.wendangku.net/doc/7d3284606.html,ing the residuals of this Probit regression,two correction terms are constructed.The?rst one is the usual Inverse Mills Ratio(IMR henceforth), which is the ratio of the probability density function and the cumulative density function evaluated at the predicted outcomes,divided by the standard error of the Probit esti-mation.This term allows controlling for the sample selec-tion bias.The second correction term accounts for the selection of?rms into export markets.In a B2C environ-ment,it is assumed that only those?rms that exceed a level of productivity will export to other countries.Hence,?rms are assumed to be heterogeneous.Since we do not have?rm level data,we follow the HMR assumption that if a?rm in country j chooses to export this is because it can at least break even in terms of pro?ts.These two addi-tional terms to control for selection into the export market and for heterogeneity in?rm-level productivity are func-tions of?tted values from the Probit and are introduced in an exponential way into the regression equation.Hence, the second step in HMR consists of the estimation of this equation using Nonlinear Least Squares.In Section5we present and compare the result of OLS and HMR estimation methods.

The interpretation of the coef?cient for the distance variable in the gravity equation is not straightforward. Apart from transport costs directly linked to geographic distance,it may also include import tariffs,costs due to regulatory differences between countries,?nancial trans-action costs,and information costs to bring the trading partners together in a transaction,etc.Since we are looking at intra-EU trade,there are no import tariffs on these cross-border transactions.The distance elasticity may also mea-sure differences in regulatory barriers,combined with ways of getting around these barriers by switching from of?ine to online trade.The introduction of a legal gover-nance quality variable may capture consumers’regulatory ‘‘regime switching’’behaviour(see below).

Since goods still need to be physically transported to the consumer following an online transaction,we can assume that transport costs remain important in online trade.Online B2C trade usually implies transport of individual small parcels while of?ine B2B may bene?t from economies of scale in large cargo consignments.Conse-quently,physical transport costs for goods bought online could actually be higher than of?ine.On the other hand, the higher number of intermediaries in of?ine trade (wholesalers,importers,etc.)may add to of?ine trade costs.We have no data to compare online and of?ine trade costs between27EU member states and therefore limit the analysis to online trade costs only.We introduce

86 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

an explicit parcel delivery cost variable in the gravity equa-tion to test the importance of physical transport costs for online trade.

The gravity equation can also handle observations on domestic trade(i=j).In that case,domestic distance is a measure of the size of a country.In line with the method-ology applied by Pacchioli(2011),McCallum(1995),Wolf (2000),we introduce a dummy variable for domestic trade observations.The coef?cient of this dummy is an indicator of home bias,or the extent of consumer preference for domestic over foreign products.The home bias factor essentially measures the combined impact of all the vari-ables that drive online(or of?ine)sales,including any omitted variables in the gravity equation such as‘‘natural’’preference for the home market.We calculate home bias only for online trade since we have no information on domestic sales for of?ine products.However,we can com-pare with home bias estimates for of?ine trade produced by other authors.

4.Data

We use data from an online consumer survey in the27 EU Member States(Civic Consulting,2011).The survey contains information on consumer online expenditure on goods only,at home as well as abroad.We use these data to construct a27?27bilateral online trade matrix for the EU27.We also construct an of?ine trade matrix between the same trading partners and for the same types of goods,so that we can compare online and of?ine trade patterns.The of?ine trade data are constructed on the basis of Comext data for the corresponding online sales product categories reported in the consumer survey.For example, when consumers report buying books or pharmaceuticals online,we use the nearest two-or four-digit CN goods cat-egory from the Comext trade database,in this case CN30‘‘Pharmaceuticals’’and CN4901‘‘printed books,brochures and similar printed materials’’to calculate the value of off-line traditional cross-border trade for these goods.Admit-tedly,these are not perfect matches but should represent a good proxy.

See Appendix A for a detailed explanation on the con-struction of the online bilateral trade matrix.

A critical issue in the construction of the online matrix is the extrapolation from survey level to population level. Aggregated at the national level,the survey data produce an estimate for average expenditure per consumer in country i on online goods in country j.We assume that the survey average is representative of online consumer behaviour in country i.We multiply this with a factor that represents the share of internet users and the share of users who actually buy online in the total population to extrapolate the survey average to the national average. We use Eurostat data for the percentage of population that is connected to the internet(see Table C1in Appendix C). However,there is a large difference between the Eurostat and the survey?gures for the share of online consumers who actually buy online and buy online abroad.Since the Eurostat?gures(43and10%of the population respec-tively)are lower than the survey?gures(63and32%respectively),we stick to Eurostat to avoid overestimation. The survey?gures would suggest that the EU Digital Agenda policy targets of getting50%of all EU consumers to buy online and20%actually shopping online abroad have already been reached in2011;the Eurostat data sug-gest otherwise.

Based on the consumer survey,we estimate the total value of online B2C trade in goods in the EU at241billion €in2011.2Out of that total,197billion€(80%)is traded domestically.Only about44billion€(18%)crosses borders between EU Member States,and another6billion€(2%)is imported from non-EU countries.

The overall?gure of241billion€looks reasonable when compared to other data sources on e-commerce.To the best of our knowledge there is no other report that esti-mates e-commerce market size for the EU as a whole. However,separate national market reports may give a good indication.For example,a Cap Gemini/IMRG report3 estimates that the UK online market,the second largest in the EU,reached82bln€in2011and91bln￡in2013.Taking into account that the UK represents about20per cent of the total EU e-commerce market in our survey data,an extrapo-lation of the UK estimate would bring total EU online market size to around410bln€.This?gure is much higher than our estimate of241bln€because it includes B2B.The ratio of 241/410=0.58suggests roughly a two-thirds/one-third split of the market between B2C and B2B.That looks plausible and is con?rmed by other sources.4Another source reports total online sales of158bln€in the combined market of France,the UK,the Netherlands,Belgium,Luxembourg and Ireland in2012.5These countries represent about40per cent of the total in our consumer survey https://www.wendangku.net/doc/7d3284606.html,ing survey data to extrapolate this?gure to EU level would imply a total EU market size of some395bln€,close to the estimate above.Finally,a third source estimates the global B2C e-commerce market at1trillion€in2012.6The EU?gure of242bln€for2011or about a quarter of the world market seems plausible.These three examples suggest that our esti-mates of the market size are plausible and consistent with other data sources.Moreover,possible biases in the scale of e-commerce should not affect the regression results and the main?ndings of the paper.

Comparing the value of estimated online cross border trade(44billion€)and observed of?ine intra-EU trade in the corresponding products categories(491billion€) (Comext),we conclude that online trade represents about 8.7%of all cross-border trade in the EU.This indicates that online orders for the relevant categories of goods consti-tute a signi?cant part of physical cross-border trade in goods.

The question arises to what extent the of?ine and online trade?gures are actually comparable.On the one

2Our consumer survey data do not include information on B2B online trade(?rms’online purchases).

3See:https://www.wendangku.net/doc/7d3284606.html,/news/uk-news/ps91-billion-spent-online-in-2013-imrg-capgemini-e-retail-sales-index.

4See:https://www.wendangku.net/doc/7d3284606.html,/2012/06/14/global-e-commerce-sales-will-top-125-trillion-2013.

5See http://www.e-commercefacts.be/research/.

6See:https://www.wendangku.net/doc/7d3284606.html,/2012/06/14/global-e-commerce-sales-will-top-125-trillion-2013.

E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–9687

hand,of?ine and online trade involve the sale of identical consumer products:books,electronics,clothing,etc.These are?nal products and the trade volume is determined by consumer demand for these goods.However,the organisa-tion of both supply chains is very different.Of?ine trade is mostly conducted business-to-business(B2B).Wholesalers export and import and use retailers as intermediaries before a good reaches the?nal consumer.By contrast, online trade is mostly B2C,with online wholesalers selling directly to?nal consumers.Differences in supply chains may,in turn,result in differences in the structure of the trade costs that underpin the two sets of trade?ows. Wholesalers often have established relations with their foreign customers,with a?xed cost that can be amortized over many transactions.Transaction size is likely to be lar-ger,again inducing economies of scale.Of?ine B2B cross-border trade?gures would have to be augmented with retail gross price margins to produce a trade value?gure that is comparable to direct B2C estimates.The above esti-mate of online B2C representing8.7%of B2B cross-border trade should therefore be interpreted with caution.

Distance estimates were obtained from CEPII.We use distance between capitals.Domestic distances are based on the greatest circle method.EU GDP?gures are taken from Eurostat.Besides the bilateral online trade estimates, we have complemented Eq.(3)with several explanatory variables that are expected to in?uence trade among coun-tries.On top of the standard Newtonian gravity variables we add three types of explanatory variables.

4.1.Cultural and institutional variables

Contemporary applications of the gravity trade model routinely include shared language between trading part-ners as an explanatory variable,and in most cases this turns out to be signi?cant.This could be considered as a proxy for‘‘cultural distance’’(Blum and Goldfarb,2006). In a B2C trading environment a shared language is essen-tial,though the relative importance of language may vary by type of good.It is likely to be more important for cross-border trade in books for instance,than for electronic goods that are more or less standardized across the world. Our dataset does not allow us to separate trade by type of good however.We also introduce a dummy for the largest and most widely shared language groups in the EU,Eng-lish,French and German,as another measure of language in?uence on cross-border e-commerce.

To measure the role of institutional quality in online trade,we construct an indicator of the quality of the legal system,based on the World Bank dataset of global gover-nance indicators.This is meant to capture the differences in expected trade costs related to dispute settlement between importers and exporters in online trade.One peculiar aspect of online B2C in the EU is that consumers buying abroad are still protected by consumer laws at home,not the law in the exporting country.This means that consumers do not really have a choice of legal regime in which they carry out their online transactions.Still,con-sumers may not be aware of this;hence,when they choose between foreign regimes,they choose the one that looks more trustworthy after comparing the quality of both legal systems.A coef?cient close to zero would indicate that consumers are aware of the legal issues.

4.2.Quality of the online enabling environment

It is important to identify possible trade costs linked to the speci?c organisational needs of online transactions in goods.Though they may be subsumed in the catch-all‘‘dis-tance’’variable we introduce three explanatory variables explicitly related to the overall enabling environment for online trade in goods.The?rst two are related to online payment systems,the third to transport costs.Since con-sumers need to have easy access to online means of cross-border payments to settle a transaction at the lowest possible transaction cost.We capture the maturity of online payment systems in two ways.First,the market share of cash payments on delivery is considered to be an indicator of the relative underdevelopment of payments systems,combined with an absence of trust in online pay-ments and high transaction costs(the transport of money). Compared to credit or debit card payment systems,it is a costly and risky system as it involves the transport of large amounts of cash,and transporter and consumer need to be available at the same location and at the same point in time.Second,the market share of PayPal is taken as a proxy of the maturity of online payment systems whereby con-sumers trust a non-bank?nancial intermediary.It may however also point to de?ciencies in the local banking sys-tem so that PayPal helps consumers to circumvent these de?ciencies.Credit and debit cards are widely available in almost every country and supported by the banking sys-tem.We do not take the share of credit and debit cards as an indicator.These cards are very common in all EU coun-tries and their share of transactions is highly negatively correlated with the previous two variables.In fact,cash-on-delivery and PayPal are also negatively correlated.To avoid multicollinearity problems we use these variables in separate regressions.Both cash-on-delivery and PayPal indicators are obtained from the World Payments Report by CapGemini et al.(2011).

An ef?cient parcel delivery system needs to be in place to physically ship the goods from their warehouses to the consumer and to minimize physical transport costs and delivery time.As argued above,the shift from of?ine to online trade only reduces the information cost component of trade costs,not the physical transport cost;on the con-trary,because of diseconomies of scale in parcel delivery compared to bulk cargo,physical transport costs may actu-ally increase.The role of transport costs on trade is not unambiguous in the literature.Lendle et al.(2012)show that the distance coef?cient is almost not affected by the inclusion of shipping costs.In this line Martinez-Zarzoso and Nowak-Lehman(2007)analyze the determinants of maritime transport and road transport costs for exports and?nd that distance is not a good proxy for transporta-tion costs.Kuwamori(2006)show that distance is impor-tant to determine transport costs,but not decisive.We capture this by introducing a parcel delivery cost indicator: the ratio of foreign to domestic parcel delivery costs,taken from Meschi et al.(2011).We take foreign parcel delivery costs by country pair and direction of trade.Parcel

88 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

transport costs are asymmetric for a given country pair. The data cover parcel delivery costs by postal services, not commercial courier services.They are of?cially reported prices,not negotiated price rates for large online retailers.Postal parcel delivery prices can be broken down in two components:costs at the sending end and costs at the receiving end.These two price components vary con-siderably across countries and are often affected by the extent of liberalisation of postal markets and competition with commercial couriers.Unfortunately,we have no independent data sources to check the consistency of these data with transport costs for commercial couriers and for major online retailers that have their own logistics networks.

5.Findings

We ran the traditional OLS(Table1)and the more sophisticated HMR(Table2)versions of the gravity model. One of the reasons to use the HMR estimator is that it deals with the zero trade?ows in the data.In our sample,120 out of1431are zero?ows,which approximately represent an8%of the sample.7From this120,119belongs to online trade(16%)and1to of?ine trade(0.14%).The results are actually very similar.We focus the discussion here on the HMR model results but will point out some differences with the OLS results,where relevant.

We start with a comparison between online and of?ine trade in a basic gravity model setting,with distance,and language as explanatory variables.We have data for both domestic and international trade in the online case,but only international trade data in the of?ine case.For the sake of comparison,we have included one regression in which we exclude domestic trade for online data on pur-pose.We use country?xed effects in these estimations. We then focus on the online gravity model and add more speci?c e-commerce related variables:governance,parcel delivery costs,online payments systems,and exporter lan-guage dummies.Finally,we estimate home bias in online trade.For each speci?cation we present two columns. The?rst shows the results from the Probit selection equa-tion that estimates the probability of two countries trad-ing.In the second column,the expected value of trade, conditional on non-zero trade between a country pair,is estimated using a Nonlinear Least Squares(NLS)regression equation.A selection variable is required to identify the parameters on both equations.This exclusion variable should affect only the decision process and should be cor-related with a country’s propensity to export but not with its current levels of exports.HMR proposes to include two indicators,regulation of entry costs and common religion. Shepotylo(2009)propose several governance indicators of regulatory quality.We have opted for the number of procedures to build a warehouse as reported by the World

Table1

OLS estimates.

Online Online,no internal

trade Of?ine Online,postal

costs

Online,

PayPal

Online,

cash

Online,home

bias

Online,language

dummies

Dep.variable log CBT log CBT log CBT log CBT log CBT log CBT log CBT log CBT

lnDistanceà0.899***à0.740***à1.349***à0.723***à0.773***à0.773***à0.639***à0.899***

[0.0812][0.0925][0.0997][0.0983][0.117][0.117][0.0955][0.0812]

Common language 2.564*** 1.315***0.657** 1.307*** 1.305*** 1.305*** 1.505*** 2.564***

[0.268][0.219][0.287][0.221][0.265][0.265][0.215][0.268]

Postal costsà0.00429

[0.0113]

PayPal0.0685***

[0.00639]

Cashà0.00927**

[0.00466]

Home bias 2.804***

[0.375]

English 4.131***

[0.515]

French 2.909***

[0.489]

German 1.510***

[0.498]

Constant11.22***10.42***15.22***10.39***10.35***10.95***9.723***9.706***

[0.598][0.643][0.702][0.642][0.794][0.794][0.660][0.749]

Observations610583701582363363610610

R-squared0.8380.8370.8780.8370.8730.8730.8570.838

Notes:Robust standard errors in brackets.

PayPal is de?ned as the logarithm of market share of PayPal in transactions in importing country.Cash is the logarithm of market share of cash-on-delivery transactions in importing country.The number of observations varies due to the number of missing data in each variable.For PayPal and Cash there are339, and for postal costs28missing observations.

***p<0.01.

**p<0.05.

*p<0.1.

7The proportion of zeros in the sample is relatively low.The reason is

that the data is aggregated at the country level ant the sample only includes

EU countries.

E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–9689

Bank.Table2shows that this variable is positive and sig-ni?cant in all Probit equations.The IMR is signi?cant as well,indicating the presence of selection bias.

There are some important changes in the coef?cients when trade moves from of?ine to online platforms.As expected,we?nd that distance matters far less online than of?ine.The coef?cient of the distance variable is sharply reduced when moving from of?ine(elasticity ofà1.294) to online(elasticity ofà0.747)trade.This con?rms the ?ndings from previous studies(Blum and Goldfarb,2006; Horta?su et al.,2009;Lendle et al.,2012).However,other sources of information-related trade costs become more prominent in online trade,in particular language.The coef-?cient for shared language between trading partners is 0.584for of?ine trade and jumps to2.320for online trade. The language coef?cient declines again when we introduce other variables in the regression but remains strongly posi-tive.In the OLS regression in Table1,the language coef?-cient increases fourfold from0.657to2.564.Lendle et al. (2012)also?nd a fourfold increase in the importance of language in their eBay study.In addition,the last column in Table2shows that when the language of the exporter country is English,French or German,there is an additional positive effect on trade.These effects are further ampli?ed in the OLS regression in Table1.English-language export-ers have a strong advantage in online markets.A plausible explanation for the role of language in cross-border trade is that in an of?ine B2B trade environment with established long-term relationships,economies of scale may facilitate the amortization of translation costs,for instance by means of translated catalogues or hiring multilingual staff to deal with foreign clients.This is more dif?cult in a B2C online trading environment where consumers have direct exchanges with e-merchants.Besides that,consumers need to be able to read the website,which in some cases may not be translated to other languages.

The coef?cient for the governance variable,the quality of the legal system,is not statistically signi?cant,so essentially not different from zero.This would lend sup-port to the objective information hypothesis.Consumers have no choice of legal regime in their online cross-border transactions;EU consumer protection rules put them under the protection of the legal regime in their home country.

Next,we introduce variables typically related to online trade.Somewhat surprisingly,the introduction of parcel delivery costs produces no signi?cant results,neither in the HMR nor in the OLS regressions.This is in line with Lendle et al.(2012)results.They?nd shipping costs to be unrelated to distance.However,online payments systems have signi?cant effects.As expected,sophisticated systems like PayPal boost cross-border e-commerce while costly systems like cash-on-delivery reduce it.

Since our dataset includes domestic online trade,we can introduce a dummy for domestic trade in the regres-sion to estimate home bias,the‘‘inherent’’preference to buy on the domestic market rather than abroad.The esti-mated coef?cient of2.775translates into a border effect in EU online markets of approximately16(exp(2.8)=16). This means that consumers are about16times more likely to buy a product on the home market than on cross-border markets.Since we do not have domestic trade values for the of?ine dataset,we can only compare this?gure with estimates for of?ine trade from the available international of?ine trade literature.Our online estimate is at the higher end of available of?ine estimates for home bias.For exam-ple,Pacchioli(2011)compares home bias in of?ine trade for EU Member States and US states.Depending on the speci?cation of the gravity model,she?nds border effects in the EU between7.4and24:EU Member States are between7and24times more likely to buy at home than in any other EU Member State,considerably higher than in the US where border effects are estimated to be between 2.6and7.One can question whether our?nding for online trade,based on a limited number of online traded con-sumer products,is comparable with the home bias values found for overall goods trade patterns,including consumer goods as well as intermediates and primary products.Ide-ally,the comparison would have to be made for a similar product composition.If not,selection bias may lead to dis-torted results.More research will be required to get a bet-ter understanding of the magnitude and sources of home bias in online trade,compared to of?ine trade.We prefer not to draw any strong conclusions from our home bias estimate.We only conclude that there is no evidence yet that points to a structural reduction in home bias in online trade.Although distance-related trade costs seem to have been reduced substantially,other trade costs have come. On balance,there is no evidence yet that total online trade costs are signi?cantly lower than of?ine trade costs.

Understanding these shifts in trade costs and the net overall effect is important for EU policy makers who want to boost online cross-border trade in a linguistically fragmented EU https://www.wendangku.net/doc/7d3284606.html,nguage is not really a policy instrument;policy makers cannot‘‘regulate’’the use of online languages.Most large e-commerce operators have multi-lingual version of their platforms.For small opera-tors this may be costly.Just to give an idea of how much drag linguistic fragmentation generates for cross-border e-commerce in the EU,we have simulated a hypothetical scenario:what would be the volume of cross-border e-commerce in the EU if the EU would have a single lan-guage,i.e.if the EU market would be like the US market? To this aim,we use Eq.(3)including PayPal variable,which correspond to column10in Table2,estimated using HMR methodology.The results show that this scenario would boost cross-border trade in the EU by9%.Although lan-guage is not a policy instrument,it is an instrumental var-iable for e-commerce platform operators.This estimate shows how important multi-lingual versions could be for e-retailers to boost their export markets.

Apart from this hypothetical exercise,we have exam-ined more realistic policy variables.Policy makers could for example promote compatibility and interoperability of online payment systems for cross-border payments. Using the same speci?cation and estimation method,we have simulated the potential impact of such an initiative by allowing for an increase in the market share of PayPal, an internationally accepted online payment system,in all countries.A1%increase in the market share of PayPal would lead to a7%increase in cross-border e-commerce. This shows that compatibility in online payment systems is

90 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

potentially a very signi?cant driver for online cross-border trade.Policy makers are also focusing on increasing the ef?ciency of parcel delivery systems,notably through enhanced competition in parcel delivery markets,as a way to promote online trade.However,since parcel deliv-ery costs is not a statistically signi?cant variable in our gravity model there is no point in running a simulation on that variable.6.Robustness checks

In this section we perform several robustness checks using the baseline speci?cation (?rst column of Table 1).First,we test the importance of unobserved heterogeneity in the data.The ?rst column of Table 3shows the results of the pooled data regressions.Though the coef?cients do not vary so much,we can observe in Fig.1that the predictions are actually quite bad.This reinforces the idea of the pres-ence of unobserved heterogeneity in data that provokes a bias in the estimation (see Fig.2).

Next,we explore to what extent the zero problem is important in our dataset.We substitute the zero ?ows by a small amount (one).By doing this transformation,the logarithm of these ?ows takes value zero,instead of being a missing value.The ?rst column of Table 3shows that the results do not change substantially.

In the third column of Table 3we include in the estima-tion some other variables that are frequently included in the gravity equation,as common border and common cur-rency.The ?rst one takes the value 1if both countries share a border,whereas the second takes value one if both countries share a currency.We expect a positive coef?cient in both cases,since both factors are supposed to facilitate trade.The results con?rm our predictions,though the effect of a common currency seems not to be signi?cant.The distance coef?cient is notably reduced when including these variables,which indicates that what distance vari-able captures is not only physical but also cultural proximity.

T a b l e 2(c o n t i n u e d )

O n l i n e ,l e g a l s y s t e m O n l i n e ,l e g a l s y s t e m O n l i n e ,c a s h O n l i n e ,c a s h O n l i n e ,h o m e b i a s O n l i n e ,h o m e b i a s O n l i n e ,l a n g u a g e d u m m i e s O n l i n e ,l a n g u a g e d u m m i e s V a r i a b l e s

P r o b i t

N L S

P r o b i t N L S P r o b i t N L S P r o b i t N L S

G e r m a n

0.793*

0.612**

[0.467][0.272]C o n s t a n t

31.02***

9.637***

38.50***

13.09***

31.21***

11.22***

30.58***

8.783***

[7.225][0.767][9.340][0.849][7.247][0.650][6.822][0.830]O b s e r v a t i o n s 432610127363432610432610L o g L i k e l i h o o d

à155.5à680.7

à35.47à255.0à155.7à599.7

à155.7

à680.8

N o t e s :R o b u s t s t a n d a r d e r r o r s i n b r a c k e t s .P a y P a l i s d e ?n e d a s t h e l o g a r i t h m o f m a r k e t s h a r e o f P a y P a l i n t r a n s a c t i o n s i n i m p o r t i n g c o u n t r y .C a s h i s t h e l o g a r i t h m o f m a r k e t s h a r e o f c a s h -o n -d e l i v e r y t r a n s a c t i o n s i n i m p o r t i n g c o u n t r y .L e g a l s y s t e m i s a d u m m y v a r i a b l e t h a t t a k e s v a l u e o n e w h e n t h e q u a l i t y o f t h e l e g a l s y s t e m i n b o t h c o u n t r i e s i s s i m i l a r .T h e n u m b e r o f o b s e r v a t i o n s v a r i e s d u e t o t h e n u m b e r o f m i s s i n g d a t a i n e a c h v a r i a b l e .F o r P a y P a l a n d C a s h t h e r e a r e 339,a n d f o r p o s t a l c o s t s 28m i s s i n g o b s e r v a t i o n s .T h e v a r i a t i o n i n t h e n u m b e r o f o b s e r v a t i o n s i n t h e p r o b i t s t a g e t a k e s p l a c e b e c a u s e a l l v a r i a b l e s t h a t p e r f e c t l y p r e d i c t s u c c e s s o r f a i l u r e i n t h e d e p e n d e n t v a r i a b l e a l o n g w i t h t h e i r a s s o c i a t e d o b s e r v a t i o n s a r e d r o p p e d .I n t h o s e c a s e s ,t h e e f f e c t i v e c o e f ?c i e n t o n t h e d r o p p e d v a r i a b l e s i s i n ?n i t y (n e g a t i v e i n ?n i t y )f o r v a r i a b l e s t h a t c o m p l e t e l y d e t e r m i n e a s u c c e s s (f a i l u r e ).D r o p p i n g t h e v a r i a b l e a n d p e r f e c t l y p r e d i c t e d o b s e r v a t i o n s h a s n o e f f e c t o n t h e l i k e l i h o o d o r e s t i m a t e s o f t h e r e m a i n i n g c o e f ?c i e n t s a n d i n c r e a s e s t h e n u m e r i c a l s t a b i l i t y o f t h e o p t i m i z a t i o n p r o c e s s .***p <0.01.**p <0.05.*p <0.1.

Table 3

Robustness checks.Variables (1)Online,log CBT (2)Online,log CBT (3)Online,log CBT log_dist

à0.614***à1.088***à0.417***[0.149][0.0998][0.113]Common language 2.926*** 2.383***0.915***[0.424]

[0.313]

[0.255]Common border 1.090***[0.212]Common currency 0.0738[0.174]Constant

9.061***12.43***7.996***[1.070][0.723][0.829]Observations 610729583R -squared

0.196

0.810

0.848

Robust standard errors in brackets.

***

p <0.01.**

p <0.05.*

p <0.1.

92 E.Gomez-Herrera et al./Information Economics and Policy 28(2014)83–96

7.Summary and conclusions

We could paraphrase Marc Twain and say that ‘‘rumours about the death of distance are greatly exagger-ated’’.Nevertheless,there is some truth in this rumour. First,the results show that the importance of geographical distance is strongly reduced in online trade,compared to of?ine trade,due to a drastic reduction in information costs in the digital economy that enables consumers to scan a much wider territory to satisfy their wishes and place their buying orders.On the other hand,there is a strong increase in the trade costs associated with crossing linguistic borders.The change in coef?cient values for dis-tance and language is con?rmed across different regression models.Second,the models that we run do not attribute any statistical signi?cance to the cost of parcel delivery in the observed patterns of cross-border e-commerce in the EU.However,the ef?ciency of online payments sys-tems is an important driver for cross-border online trade in the EU.This leaves policy makers with little regulatory margin to boost cross-border online trade.The data only demonstrate that improvements in compatibility and interoperability between online payment systems would be a step in the right direction.Third,the results provide a preliminary indication that home bias is not signi?cantly different in online markets compared to traditional of?ine trade.Despite the fact that reduced information costs widen the market for consumers and facilitate buying abroad,consumers still have a strong tendency to buy at https://www.wendangku.net/doc/7d3284606.html,nguage barriers certainly play an important role here,but other as yet unobserved variables may also be part of the explanation.

EU policy makers have?xed Digital Agenda policy tar-gets for e-commerce in terms of increasing volumes of online(cross-border)trade.This might be surprising because trade is only a means to enhance consumer wel-fare,not an end in itself.E-commerce can boost consumer welfare through lower transaction costs,increased diver-sity of supply and more price competition.Our data do not allow an investigation of these welfare effects though we can assume that the volume of online trade is a good proxy indicator of consumers’perceived bene?ts.In that sense,e-commerce policy follows in the footsteps of the EU of?ine Single Market that aims to reduce trade barriers and boost cross-border trade with a view to stimulate price competition and increase the diversity of supply. E-commerce dramatically reduces the transport cost of information.This opens up a much wider geographical catchment area for consumers and suppliers.This paper shows however that(cross-border)e-commerce is still subject to trade barriers;not only in terms of physical delivery costs and regulatory barriers but also new trade costs induced by linguistic market segmentation and online payments systems.

The total volume of consumer online expenditure is likely to increase over time as more consumers become more con?dent with online shopping and move a larger share of their shopping online.An important limit on that growth potential is the composition of the online shopping basket.The consumer survey data that we use show that this is heavily biased towards a limited number of goods such as electronics,clothing,music/?lm and a few other items.The online shopping basket differs considerably from the overall consumer goods basket,probably because other types of goods do not lend themselves so easily to online trade.Further research is also needed to explain the composition and restrictions on the online consumer basket and explore ways to widen the range of goods that can be traded online.Even if the total volume of online shopping still has very considerable growth potential,the gravity model indicates that the ratio of domestic to for-eign online shopping may not change that much because it is held back by linguistic fragmentation in the EU market. Since only36out of729EU27country pairs share a com-mon language,online retailers who want to expand their business abroad are strongly advised to have a range of language versions of their websites.However,it is dif?cult to see how language could become an instrumental vari-able for policy makers.

A?nal word of caution.This analysis is based on a sin-gle EU consumer survey data set that offers some unique insights into the value and direction of online cross-border trade between EU countries.Obviously,these data do not have the same validity as the far more comprehensive and detailed international of?ine trade in goods statistics

E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–9693

that have accumulated over the years.They offer a?rst insight but more effort will have to go into the construc-tion of more comprehensive and reliable online cross-border trade data sets that will enable a more detailed and rigorous testing of the drivers and impediments to online cross-border trade.Further work would have to include more details on product-speci?c cross-border trade,transport costs,prices and information costs. Appendix A.The construction of the online bilateral trade matrix

For the purpose of this article we use a unique dataset from an online survey of29.100consumers in the27Mem-ber States of the EU,carried out by Civic Consulting(2011) on behalf of the European Commission.The advantage of consumer survey data is that they offer a more compre-hensive picture,not affected by speci?c market and prod-uct biases of company data.The disadvantage of this survey is that it generates no details on products and prices and only provides an overall expenditure pattern.Eurostat household and?rm level survey data also offer some insights into consumer online spending and company online sales.However,they provide no information on the geographical direction of cross-border online trade; consequently,they could not be used for the purpose of this research.Another major advantage of the consumer survey is that it covers online trade in goods,a subject on which there is little information to date.At the same time, the fact that it covers goods only constitutes a major limi-tation.Online trade in services is probably more important than online goods trade.

The survey also makes an explicit distinction between online expenditures on domestic and foreign websites. One may question if consumers are consistently able to distinguish between these categories.The dot com or country extensions of web addresses are not always a good indicator of the actual geographical location of the sup-plier.Some major e-merchants have physical supply net-works that are unrelated to the website addresses.That opens up several possibilities for the de?nition of cross-border online trade in goods.The simplest de?nition is an online transaction that triggers a?ow of goods crossing one or more national borders.However,these may not necessarily be the same border(s)as between the country of residence of the buyer and the country of origin of the website of the seller–depending on where the warehouse of the seller is located.Another possible de?nition is a transaction that triggers a?nancial transfer across national borders,independently of the underlying physical transac-tion.In this study we stick to the simple de?nition and assume that the physical delivery and the?nancial trans-action follow the same geographical pattern.In the under-lying consumer survey data it is assumed that cross-border online trade means that the good crosses at least one state border.

Questionnaires were administered through computer-assisted web interviews.As such,they cover consumers who have internet access and could potentially carry out e-commerce transactions,though not all of them actually do so.Overall,85%of the respondents in the sample are online shoppers,i.e.they bought at least one product online in the twelve months prior to the survey date.This compares to an EU average of52%of respondents who have Internet access at home and who do online shopping, according to a recent Eurobarometer(2011)survey.Online panels may over-represent online shoppers,though there are no a priori reasons to give more credibility to either of these surveys.It is possible that online panels are biased towards consumers who feel more at ease with computers and the internet.Oversampling is not problematic when the data are used to analyze the patterns of domestic and foreign online shopping.We should however bear the potential risks of oversampling in mind when we extrapo-late the survey data to population levels to estimate the extent and total value of online shopping.We use the more conservative Eurobarometer(2011)data to extrapolate,to avoid overestimation.Panel size was approximately1000 per country,with some variation according to country size. Country survey panels were built to ensure that all key demographic groups(e.g.gender,age,region,household size,occupation)are represented.The sample distribution for gender and age is close to the national?gures.

The survey questionnaire contains information about the number of domestic and foreign online transactions over the last12months,the countries where cross-border transactions were made,and the amount of money spent on domestic and foreign transactions.We use this infor-mation to construct a?rst matrix of the sample-level value of online transactions among the27EU Member States.Theoretically,the matrix could contain up to 27?27=729trade observations.In practice,some cells are empty when no cross-border transactions are reported for particular pairs of countries.The survey also contains information on cross-border transaction with non-EU countries:the US,China,Norway,Iceland and Switzerland,and the residual category Rest of the World. The diagonal line of that matrix contains the value of domestic online transactions.For the non-diagonal cells (cross-border trade),we use survey information on the total amount spent per consumer on cross-border trans-actions over the last12months,and the countries where this spending took place.We calculate average spent for each consumer per cross-border transaction and apply the same average to all transactions,assuming that all cross-border transactions for a given consumer have the same value.This is admittedly a simpli?cation but survey data do not allow us to be more speci?c.

Since all country survey samples have a more or less similar size of1000respondents,the sample-level trade matrix is not representative of the population-level trade pattern in the EU.10%of respondents in Malta doing a transaction with the rest of the EU have not the same economic weight as10%of German respondents doing a cross-border transaction.We therefore construct a second trade matrix at population level,using the ratio of sample (online)population to total(online)population as a multiplier.Total online population?gures are taken from Eurostat.

94 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

Appendix B

See Table B1.

Appendix C

See Table C1.References

Anderson,J.,Van Wincoop,E.,2003.Gravity with gravitas:a solution to the border puzzle.Am.Econ.Rev.93,170–192.

Baier,S.L.,Bergstrand,J.H.,2009.Estimating the effects of free trade agreements on international trade?ows using matching econometrics.J.Int.Econ.77,63–76.

Berthelon,M.,Freund, C.,2008.On the conservation of distance in international trade.J.Int.Econ.75,310–320.

Blum,B.,Goldfarb,A.,2006.Does the internet defy the law of gravity.J.

Int.Econ.70,384–405.

Cairncross,F.,1997.The Death of Distance;How the Communications Revolution will Change Our Lives.Orion Business Books,London. CapGemini,Royal Bank of Scotland,EFMA,2011.World Payments Report,

WPR_2011.pdf>.

Civic Consulting,2011.Consumer market study on the functioning of e-commerce and internet marketing and selling techniques in the retail of goods..

Coughlin,C.,Novy,D.,2009.Is the International Border Effect Larger than the Domestic Border Effect:Evidence from US Trade,Research Article Series2009/29.Nottingham University,Leverhulme Centre. Disdier,A.,Head,K.,2008.The puzzling persistence of the distance effect on bilateral trade.Rev.Econ.Stat.90,37–48.

Eurobarometer,2011.Special report on consumer empowerment.Report No.342,.

European Commission,2012.A coherent framework to build trust in the Digital Single Market for e-commerce and online services,

communication_2012_en.htm>.

Feenstra,R.C.,2002.Border effects and the gravity equation:consistent methods for estimation.Scot.J.Polit.Econ.49,491–506.

Frankel,J.,Wei,S.J.,1995.Trading blocks and the Americas.J.Dev.Econ.

47,61–96.

Table B1

Frequency distribution by type of good in the sample survey.Source:Civic

Consulting(2011)and own calculations.

Type of good%

Electronics19

Cloth and shoes17

Books10

Music/video6

Cosmetics6

Software6

Electrical6

Toys5

Sports Eq.4

Car parts3

Furniture2

Tools2

Medicines2

Other12

Total100

Table C1

Percentage of population buying online and abroad.Source:Eurostat for population and percentage online,online buyers and buyers abroad from the Civic Consulting(2011)consumer survey and from Eurostat.

Country Population

(0000)%

Online

%Online

that buys

%Online that

buys abroad

%Pop that buys

online/survey

%Pop that buys

online abroad/

survey

%Pop that buys

online/eurostat

%Pop that buys

online abroad/

eurostat

Austria8.37280967776624432

Belgium10.82783845169434324

Bulgaria7.576517234371773

Cyprus80158575633322118

Czech10.5127396247018305

Denmark 5.54791955586507028

Estonia 1.34077722656202110

Finland 5.35089945284476228

France64.70980944075325314

Germany81.7578397418134649

Greece11.1255391574830187

Hungary10.0137069154810224

Ireland 4.46877897468574322

Italy60.3975786504928155

Latvia 2.2497270385027208

Lithuania 3.3296574324821165

Luxemburg50291827575696556

Malta41669595841404538

Netherlands16.57792892982276914

Poland38.1646595276218302

Portugal10.6375881434725187

Romania21.46644791935961

Slovakia 5.42478964775363111

Slovenia 2.05469793654253711

Spain47.1506984465832279

Sweden9.34894954389407116

UK62.04287974584397110

Total%74854463324310

Total#502.152

E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–9695

Freund,C.,Weinhold,D.,2000.The effect of the internet on international trade.J.Int.Econ.62,171–189.

Helpman, E.,Melitz,M.,Rubinstein,Y.,2008.Estimating trade?ows: trading partners and trading volumes.Q.J.Econ.123,441–487. Horta?su,A.,Martinez-Jerez,F.,Douglas,J.,2009.The geography of trade in online transactions:evidence from eBay and Mercado Libre.Am.

Econ.J.1,53–74.

Kuwamori,H.,2006.The role of distance in determining international transport costs:evidence from Philippine import data.IDE Discussion Papers60.Japan External Trade Organisation.

Lendle, A.,Olarreaga,M.,Schropp,S.,Vezina,P.L.,2012.There goes gravity:how eBay reduces trade costs.CEPR discussion article9094. Martinez-Zarzoso,I.,Nowak-Lehmann,D.,2007.Is distance a good proxy for transport costs?The case of competing transport modes.J.Int.

Trade and Econ Dev.16,411–434.McCallum,J.,1995.National borders matter:Canada–US regional trade patterns.Am.Econ.Rev.85,615–623.

Meschi,M.,Irving,T.,Gillespie,M.,2011.Intra-Community Cross-Border Parcel Delivery.FTI Consulting,London.

Pacchioli,C.,2011.Is the EU internal market suffering from an integration de?cit?Estimating the home bias effect.CEPS working document348. Rauch,J.,https://www.wendangku.net/doc/7d3284606.html,works versus markets in international trade.J.Int.

Econ.48,7–35.

Santos Silva,J.M.C.,Tenreyro,S.,2006.The log of gravity.Rev.Econ.Stat.

88,641–658.

Shepotylo,O.,2009.Gravity with zeros:estimating trade potential of CIS countries.Kyiv School of Economics Discussion Papers16.

Wolf,H.,2000.International home bias in trade.Rev.Econ.Stat.84,555–563.

96 E.Gomez-Herrera et al./Information Economics and Policy28(2014)83–96

批处理命令行for语句

for语句可以在命令行提示符中使用，也可以在批处理文件中使用。这两种情况下唯一的区别是%和%%，参加下文说明。 一、for语句的格式： for [参数] 变量in (集合) do 命令[命令的参数] 二、for语句的作用：对集合内的元素逐一执行后面的命令。 1、如：for %%i in (你好) do echo %%i 将在屏幕上显示“你好”2个字。这里集合是“你好”，执行的命令是“echo”。由于集合中只有1个元素，因此循环只运行一次。 如果改成for %%i in (你好朋友) do echo %%i 将会显示2行文字，第一行为“你好”，第二行为“朋友”。因为2个词之间有空格，因此集合中就有了2个元素，循环将运行2次。 2、注意：以上for语句的运行方式是新建一个批处理文件，即扩展名为“.bat”的文件，内容为上面的命令，然后运行。为了批处理执行完不退出，可在最后加上一条pause>null命令，这样能看到执行的结果。要想通过cmd命令行执行的话，必须将%%换成%，即去掉一个%，如下： for %i in (你好) do echo %i 3、以下所有例子都是这样，若要在命令行提示符下执行，请将所有的%%改成一个%。 三、for语句详细说明： 上面语句格式中有的加了中括号[]，表示这个语句元素不是必须的，只在需要时使用。像刚才显示“你好”的命令中就没有使用[参数]这

个语句元素。所有语句元素间用空格隔开。 各语句元素的写法：for、in、do这3个符号是固定不变的 1、[参数]的种类：只有4种，分别是/d、/r、/l、/f（即目录Directory、递归recursion、序列list、文件file），他们用于对后面的集合的含义做出解释，请与下面的集合解释结合来看。这4个参数不区分大小写，可以联合使用，即一条for语句可以出现多个参数。 2、变量：除10个数字外（0－9）的所有符号（因为0－9往往作为形参使用，为了与此区别），变量名一般用单个字母表示即可，而且变量名区分大小写，即A和a是两个不同的变量。变量名前面必须是%，当在命令提示符下执行时，只用一个%；而在批处理程序中，必须用%%。 一行语句中，一般只需定义一个变量。使用/f参数中的tokens 选项可以将集合中的元素分解成多个值，并自动定义新的变量对应这些值。这时语句中可以使用多个变量，通常按字母顺序命名，即第一个是%%a，那么后一个就用%%b。如果第一个是%%i，后一个就用%%j。依此类推。具体看后面的相关内容。 变量可以直接在do后面的命令中使用。每次使用的变量总数不超过52个。 3、集合：集合必须放在括号里。集合是一行文本，这行文本可能有几种类型，如“你好”只是一串字符；“c:\boot.ini”是一个文件；“dir /b”是一个命令。 （1）如果for语句中没有使用任何参数，对待集合文本的处理方式是：

批处理命令for语句基本用法

批处理命令for语句的基本用法 [系列教程]批处理for语句从入门到精通[20101225更新] ____________________________版主提醒 ____________________________ 文档来自于网络搜索 为了避免影响技术讨论、提高看帖的舒适性，请大家不要在此帖下跟 无实质内容的口水帖，特别是纯顶、纯支持、纯感谢、路过之类的帖子， 管理人员将不定期清理此类回帖，请大家多参与讨论少灌水，与人方便， 终将给自己带来方便，谢谢合作。 ________________________________________________________________ 文档来自于网络搜索 批处理是一门简单的脚本语言，虽然不能独当一面，但是，若作为工作中的辅助工具，绝对会让大家有随用随写、称心如意的畅快感。 文档来自于网络搜索 和其他语言相比，批处理语言有其先天性的优势： 1、系统自带，无需另行安装； 2、命令少，语句简洁，上手非常快； 3、编写出来的脚本小巧玲珑，随写随用； 但是，因为它以命令行方式工作，操作多有不便，在图形界面大行其道的windows世界里，多多少少会让大众望而却步；就算是对命令行有好感的新手，面对微软有如天书的帮助文件，很多人也会败下阵来，因此，论坛里很多会员也发出了编写系统的批处理教程的呼声。

文档来自于网络搜索 编写系统的批处理新手教程，一直是论坛管理层讨论的热点问题，但是，各位管理人员大多都有工作在身，而系统的教程涉及的面是如此之广，面对如此浩大的工程，仅凭一两个人的力量，是难以做好的，因此，本人退而求其次，此次发布的教程，以专题的形式编写，日后人手渐多之后，再考虑组织人力编写全面的教程。 文档来自于网络搜索之所以选择最难的for，一是觉得for最为强大，是大多数人最希望掌握的；二是若写其他命令教程，如果没有for的基础，展开来讲解会无从下手；三是for也是批处理中最复杂最难掌握的语句，把它攻克了，批处理的学习将会一片坦途。 文档来自于网络搜索 这次的for语句系列教程，打算按照for语句的5种句式逐一展开，在讲解for/f的时候，会穿插讲解批处理中一个最为关键、也是新手最容易犯错的概念：变量延迟，大纲如下： 文档来自于网络搜索一前言 二for语句的基本用法 三for /f（含变量延迟） 四for /r 五for /d 六for /l 遵照yibantiaokuan的建议，在顶楼放出此教程的txt版本、word版本和pdf版本，以方便那些离线浏览的会员。 文档来自于网络搜索[本帖最后由namejm于2010-12-26 02:36编辑]

实用批处理(bat)教程

目录 第一章批处理基础 第一节常用批处理内部命令简介 1、REM 和:: 2、ECHO 和@ 3、PAUSE 4、ERRORLEVEL 5、TITLE 6、COLOR 7、mode 配置系统设备 8、GOTO 和: 9、FIND 10、START 11、assoc 和ftype 12、pushd 和popd 13、CALL 14、shift 15、IF 16、setlocal 与变量延迟(ENABLEDELAYEDEXPANSION / DISABLEDELAYEDEXPANSION 启动或停用延缓环境变量扩展名。) 17、ATTRIB显示或更改文件属性 第二节常用特殊符号 1、@命令行回显屏蔽符 2、%批处理变量引导符 3、> 重定向符 4、>>重定向符 5、<、>、<& 重定向符 6、|命令管道符 7、^转义字符 8、组合命令 9、& 组合命令 10、||组合命令 11、\"\"字符串界定符 12、, 逗号 13、; 分号 14、() 括号 15、! 感叹号 第二章FOR命令详解 一、基本格式 二、参数/d仅为目录 三、参数/R递归（文件名） 四、参数/L迭代数值范围 五、参数/F迭代及文件解析 第三章FOR命令中的变量

一、~I- 删除任何引号(\")，扩展%I 二、%~fI- 将%I 扩展到一个完全合格的路径名 三、%~dI- 仅将%I 扩展到一个驱动器号 四、%~pI- 仅将%I 扩展到一个路径 五、%~nI- 仅将%I 扩展到一个文件名 六、%~xI- 仅将%I 扩展到一个文件扩展名 七、%~sI- 扩展的路径只含有短名 八、%~aI- 将%I 扩展到文件的文件属性 九、%~tI- 将%I 扩展到文件的日期/时间 十、%~zI- 将%I 扩展到文件的大小 十一、%~$PATH:I 第四章批处理中的变量 一、系统变量 二、自定义变量 第五章set命令详解 一、用set命令设置自定义变量 二、用set命令进行简单计算 三、用set命令进行字符串处理 1、字符串替换 2、字符串截取 第六章if命令讲解 第一种用法：IF [NOT] ERRORLEVEL number command 第二种用法：IF [NOT] string1==string2 command 第三种用法：IF [NOT] EXIST filename command 第四种用法：IF增强的用法 第七章DOS编程高级技巧 一、界面设计 二、if…else…条件语句 三、循环语句 四、子程序 五、用ftp命令实现自动下载 六、用7-ZIP实现命令行压缩和解压功能 七、调用VBScript程序 八、将批处理转化为可执行文件 九、时间延迟 1、利用ping命令延时 2、利用for命令延时 3、利用vbs延迟函数，精确度毫秒，误差1000毫秒内 4、仅用批处理命令实现任意时间延迟，精确度10毫秒，误差50毫秒内 十、模拟进度条 十一、特殊字符的输入及应用 十二、随机数（%random%）的应用技巧 十三、变量嵌套与命令嵌套 1、更正了所有的错别字，适当排版，增加条理性。

Microsoft OLE DB Provider for ODBC Drivers (0x80004005)解决办法

IIS本地服务器设置调试问题集锦 Microsoft OLE DB Provider for ODBC Drivers (0x80004005)解决办法 错误类型： Microsoft OLE DB Provider for ODBC Drivers (0x80004005) [Microsoft][ODBC Microsoft Access Driver]常见错误不能打开注 册表关键字'Temporary (volatile) Jet DSN for process 0xd30 Thread 0x29c DBC 0x1762e54 Jet'。/admin_login/data_base /conn.asp, 第8 行 下面是conn.asp的全部代码： <% Dim conn,connstr,DBPath,filePath,dbName dbName = "DB.mdb" filePath = Request.ServerVariables("APPL_PHYSICAL_PATH") DBPath =filePath&"admin_login\data_base\"&dbName Set conn = Server.CreateObject("ADODB.Connection") connstr="driver={Microsoft Access Driver (*.mdb)};dbq=" & DBPath conn.Open connstr %> 以为第八行有什么问题的，看来看去就那几个字，没啥问题啊。 百度了N久才查到准确答案，记录下来以备日后使用 找到这个文件夹c:\Windows\Temp 然后右键—安全—添加Everyone用户并设置“完全控制”权限 如果没看到“安全”选项： 双击“我的电脑”—工具—文件夹选项-查看-把：使用简单文件共享（推荐）前面的勾去掉

批处理命令格式

批处理命令格式.txt人永远不知道谁哪次不经意的跟你说了再见之后就真的再也不见了。一分钟有多长？这要看你是蹲在厕所里面，还是等在厕所外面……echo 表示显示此命令后的字符 echo off 表示在此语句后所有运行的命令都不显示命令行本身 @与echo off相象，但它是加在每个命令行的最前面，表示运行时不显示这一行的命令行（只能影响当前行）。 call 调用另一个批处理文件（如果不用call而直接调用别的批处理文件，那么执行完那个批处理文件后将无法返回当前文件并执行当前文件的后续命令）。 pause 运行此句会暂停批处理的执行并在屏幕上显示Press any key to continue...的提示，等待用户按任意键后继续 rem 表示此命令后的字符为解释行（注释），不执行，只是给自己今后参考用的（相当于程序中的注释）。 例1：用edit编辑a.bat文件，输入下列内容后存盘为c: a.bat，执行该批处理文件后可实现：将根目录中所有文件写入 a.txt中，启动UCDOS，进入WPS等功能。 批处理文件的内容为: 命令注释： @echo off 不显示后续命令行及当前命令行 dir c: *.* >a.txt 将c盘文件列表写入a.txt call c: ucdos ucdos.bat 调用ucdos echo 你好显示"你好" pause 暂停,等待按键继续 rem 准备运行wps 注释：准备运行wps cd ucdos 进入ucdos目录 wps 运行wps 批处理文件的参数 批处理文件还可以像C语言的函数一样使用参数（相当于DOS命令的命令行参数），这需要用到一个参数表示符“%”。 %[1-9]表示参数，参数是指在运行批处理文件时在文件名后加的以空格（或者Tab）分隔的字符串。变量可以从%0到%9，%0表示批处理命令本身，其它参数字符串用%1到%9顺序表示。例2：C:根目录下有一批处理文件名为f.bat，内容为： @echo off format %1 如果执行C: >f a: 那么在执行f.bat时，%1就表示a:，这样format %1就相当于format a:，于是上面的命令运行时实际执行的是format a: 例3：C:根目录下一批处理文件名为t.bat，内容为: @echo off type %1 type %2 那么运行C: >t a.txt b.txt %1 : 表示a.txt %2 : 表示b.txt 于是上面的命令将顺序地显示a.txt和b.txt文件的内容。 特殊命令 if goto choice for是批处理文件中比较高级的命令，如果这几个你用得很熟练，你就是批

电脑驱动程序安装顺序

台式机和笔记本的驱动程序安装顺序 下表概述了在戴尔台式机和笔记本上安装驱动程序的正确顺序。重新安装Microsoft? Windows? 后，请按照列出的顺序来重新安装驱动程序。建议您打印此列表以供安装驱动程序时参考。 注：如果未按顺序安装驱动程序，某些设备可能无法正常工作。 1. 台式机系统软件 (DSS) 或笔记本系统软件 (NSS)—可为操作系统提供关键更新和补丁程序的重要实用程序。如果要重新安装 Windows 驱动程序或更新全部驱动程序，则一定要先安装该软件。 DSS和NSS位于 Drivers and Downloads (驱动程序和下载)页面上的System Utilities Category (系统公用程序类别)的栏目内。 2. 芯片组—帮助 Windows 控制系统主板组件和控制器。位于 Drivers and Downloads (驱动程序和下载)页面上的 ChipsetCategory(芯片组及智能卡类别)的栏目内。 3. 显卡—增强视频性能。位于 Drivers and Downloads (驱动程序和下载) 页面上的Video Category (显卡类别)的栏目内。 4. 集成网卡(NIC)—启用和增强用于互联网或有线网络访问的网络控制器。此类驱动程序位于Drivers and Downloads (驱动程序和下载)页面上的Network Category (网卡无线及蓝牙类别)的栏目内。 5. 仅限笔记电脑：Dell Quickset 或 Dell Control Point Manager (DCP)—控制笔记本电脑上的电源管理、环境光线传感器、无线配置文件和安全功能。此类驱动程序位于 Drivers and Downloads (驱动程序和下载)页面上的Applications (应用程序)的栏目内。 6. 声卡—启用和增强音频控制器。此类驱动程序位于 Drivers and Downloads (驱动程序和下载)页面上的Audio Category (声卡类别)的栏目内。 7. 调制解调器—具有拨号上网功能。此类驱动程序位于 Drivers and Downloads (驱动程序和下载)页面上的 Communication Category (Modem和通讯类别)的栏目内。 8. 无线网卡—启用和增强无线网络控制器。此类驱动程序位于 Drivers and Downloads (驱动程序和下载)页面上的Network Category (网卡无线及蓝牙)的栏目内。

批处理基础知识

批处理文件基础知识 一、单符号message指定让MS-DOS在屏幕上显示的正文 ~ ①在for中表示使用增强的变量扩展。 ②在%var:~n,m%中表示使用扩展环境变量指定位置的字符串。 ③在set/a中表示一元运算符，将操作数按位取反。 ! ①在set /a中一元运算符，表示逻辑非。比如set /a a=!0，这时a就表示逻辑1。 @ ①隐藏命令行本身的回显，常用于批处理中。 % ①在set /a中的二元运算符，表示算术取余。 ②命令行环境下，在for命令in前，后面接一个字符（可以是字母、数字或者一些特定字符），表示指定一个循环或者遍历指标变量。 ③批处理中，后接一个数字表示引用本批处理当前执行时的指定的参数。 ④其它情况下，%将会被脱去（批处理）或保留（命令行） ^ ①取消特定字符的转义作用，比如& | > < ! "等，但不包括%。比如要在屏幕显示一些特殊的字符，比如> >> | ^ &等符号时，就可以在其前面加一个^符号来显示这个^后面的字符了，^^就是显示一个^，^|就是显示一个|字符了; ②在set/a中的二元运算符，表示按位异或。 ③在findstr/r的[]中表示不匹配指定的字符集。 & ①命令连接字符。比如我要在一行文本上同时执行两个命令，就可以用&命令连接这两个命令。 ②在set/a中是按位与。 : ①标签定位符，表示其后的字符串为以标签，可以作为goto命令的作用对象。比如在批处理文件里面定义了一个":begin"标签，用"goto begin"命令就可以转到":begin"标签后面来执行批处理命令了。 ②在%var:string1=string2%中分隔变量名和被替换字串关系。 | ①管道符，就是将上一个命令的输出，作为下一个命令的输入."dir /a/b |more"就可以逐屏的显示dir命令所输出的信息。 ②在set/a中的二元运算符，表示按位或。 ③在帮助文档中表示其前后两个开关、选项或参数是二选一的。 / ①表示其后的字符（串）是命令的功能开关（选项）。比如"dir /s/b/a-d"表示"dir"命令指定的不同的参数。 ②在set/a中表示除法。 > ①命令重定向符，将其前面的命令的输出结果重新定向到其后面的设备中去，后面的设备中的内容被覆盖。比如可以用"dir > lxmxn.txt"将"dir"命令的结果输出到"lxmxn.txt"这个文本文件中去。 ②在findstr/r中表示匹配单词的右边界，需要配合转义字符\使用。 < ①将其后面的文件的内容作为其前面命令的输入。 ②在findstr/r中表示匹配单词的左边界，需要配合转义字符\使用。 . ①在路径的\后紧跟或者单独出现时：

DOS批处理命令大全

写批处理 扩展名是bat(在nt/2000/xp/2003下也可以是cmd)的文件就是批处理文件。 ==== willsort 编注======================================= .bat是dos下的批处理文件 .cmd是nt内核命令行环境的另一种批处理文件 从更广义的角度来看，unix的shell脚本以及其它操作系统甚至应用程序中由外壳进行解释执行的文本，都具有与批处理文件十分相似的作用，而且同样是由专用解释器以行为单位解释执行，这种文本形式更通用的称谓是脚本语言。所以从某个程度分析，batch, unix shell, awk, basic, perl 等脚本语言都是一样的，只不过应用的范围和解释的平台各有不同而已。甚至有些应用程序仍然沿用批处理这一称呼，而其内容和扩展名与dos的批处理却又完全不同。 =================================== 首先批处理文件是一个文本文件，这个文件的每一行都是一条DOS命令（大部分时候就好象我们在DOS 提示符下执行的命令行一样），你可以使用DOS下的Edit或者Windows的记事本(notepad)等任何文本文件编辑工具创建和修改批处理文件。 ==== willsort 题注=================== 批处理文件中完全可以使用非dos命令，甚至可以使用不具有可执行特性的普通数据性文件，这缘于wind ows系统这个新型解释平台的涉入，使得批处理的应用越来越"边缘化"。所以我们讨论的批处理应该限定在dos环境或者命令行环境中，否则很多观念和设定都需要做比较大的变动。 ======================== 其次，批处理文件是一种简单的程序，可以通过条件语句(if)和流程控制语句(goto)来控制命令运行的流程，在批处理中也可以使用循环语句(for)来循环执行一条命令。当然，批处理文件的编程能力与C语言等编程语句比起来是十分有限的，也是十分不规范的。批处理的程序语句就是一条条的DOS命令（包括内部命令和外部命令），而批处理的能力主要取决于你所使用的命令。 ==== willsort 编注================== 批处理文件(batch file)也可以称之为批处理程序(batch program)，这一点与编译型语言有所不同，就c语言来说，扩展名为c或者cpp的文件可以称之为c语言文件或者c语言源代码，但只有编译连接后的exe 文件才可以称之为c语言程序。因为批处理文件本身既具有文本的可读性，又具有程序的可执行性，这些称谓的界限是比较模糊的。 =========================== 第三，每个编写好的批处理文件都相当于一个DOS的外部命令，你可以把它所在的目录放到你的DOS搜索路径(path)中来使得它可以在任意位置运行。一个良好的习惯是在硬盘上建立一个bat或者batch目录（例如C:\BATCH），然后将所有你编写的批处理文件放到该目录中，这样只要在path中设置上c:\batch，你就可以在任意位置运行所有你编写的批处理程序。 ==== willsort 编注===== 纯以dos系统而言，可执行程序大约可以细分为五类，依照执行优先级由高到低排列分别是：DOSKEY宏命令（预先驻留内存），https://www.wendangku.net/doc/7d3284606.html,中的内部命令（根据内存的环境随时进驻内存），以com为扩

目前为止最全的批处理教程

目录 第一章 批处理基础 第一节 常用批处理内部命令简介 1、REM 和 :: 2、ECHO 和 @ 3、PAUSE 4、ERRORLEVEL 5、TITLE 6、COLOR 7、mode 配置系统设备 8、GOTO 和 : 9、FIND 10、START 11、assoc 和 ftype 12、pushd 和 popd 13、CALL 14、shift 15、IF 16、setlocal 与 变量延迟(ENABLEDELAYEDEXPANSION / DISABLEDELAYEDEXPANSION 启动或停用延缓环境变量扩展名。) 17、ATTRIB显示或更改文件属性 第二节 常用特殊符号

1、@命令行回显屏蔽符 2、%批处理变量引导符 3、> 重定向符 4、>>重定向符 5、<、>、<& 重定向符 6、|命令管道符 7、^转义字符 8、组合命令 9、& 组合命令 10、||组合命令 11、\"\"字符串界定符 12、, 逗号 13、; 分号 14、() 括号 15、! 感叹号 第二章 FOR命令详解 一、基本格式 二、参数 /d仅为目录 三、参数 /R递归（文件名） 四、参数 /L迭代数值范围 五、参数 /F迭代及文件解析 第三章 FOR命令中的变量

一、 ~I- 删除任何引号(\")，扩展 %I 二、 %~fI- 将 %I 扩展到一个完全合格的路径名 三、 %~dI- 仅将 %I 扩展到一个驱动器号 四、 %~pI- 仅将 %I 扩展到一个路径 五、 %~nI- 仅将 %I 扩展到一个文件名 六、 %~xI- 仅将 %I 扩展到一个文件扩展名 七、 %~sI- 扩展的路径只含有短名 八、 %~aI- 将 %I 扩展到文件的文件属性 九、 %~tI- 将 %I 扩展到文件的日期/时间 十、 %~zI- 将 %I 扩展到文件的大小 十一、 %~$PATH:I 第四章 批处理中的变量 一、系统变量 二、自定义变量 第五章 set命令详解 一、用set命令设置自定义变量 二、用set命令进行简单计算 三、用set命令进行字符串处理 1、字符串替换 2、字符串截取 第六章 if命令讲解 第一种用法：IF [NOT] ERRORLEVEL number command

windows批处理命令详解及脚本实例

批处理文件是将一系列命令按一定的顺序集合为一个可执行的文本文件，其扩展名为BAT。第一部分：批处理内部命令 1、REM REM 是个注释命令一般是用来给程序加上注解的，该命令后的内容在程序执行的时候将不会被显示和执行。例： REM 你现在看到的就是注解，这一句将不会被执行。在以后的例子中解释的内容都REM 会放在REM后面。请大家注意。 2、ECHO ECHO 是一个回显命令主要参数有OFF和ON,一般用ECHO message来显示一个特定的消息。例： Echo off Rem 以上代表关闭回显即不显示所执行的命令 Echo 这个就是消息。 Rem 以上代表显示"这就是消息"这列字符 执行结果： C:\>ECHO.BAT 这个就是消息。 3、GOTO GOTO 即为跳转的意思。在批处理中允许以"：XXX"来构建一个标号然后用GOTO ：标号直接来执行标号后的命令。例 :LABEL REM 上面就是名为LABEL的标号。 DIR C:\ DIR D:\ GOTO LABEL REM 以上程序跳转标号LABEL处继续执行。 4、CALL CALL 命令可以在批处理执行过程中调用另一个批处理，当另一个批处理执行完后再继续

执行原来的批处理。例： 批处理2.BAT内容如下： ECHO 这就是2的内容 批处理1.BAT内容如下： ECHO 这是1的内容 CALL 2.BAT ECHO 1和2的内容全部显示完成 执行结果如下： C:\>1.BAT 这是1的内容 这就是2的内容 1和2的内容全部显示完成 5、PAUSE PAUSE 停止系统命令的执行并显示下面的内容。例： C:\> PAUSE 请按任意键继续. . . 6、IF IF 条件判断语句，语法格式如下： IF [NOT] ERRORLEVEL number command IF [NOT] string1==string2 command IF [NOT] EXIST filename command 说明： [NOT] 将返回的结果取反值即"如果没有"的意思。 ERRORLEVEL 是命令执行完成后返回的退出值 Number 退出值的数字取值范围0~255。判断时值的排列顺序应该又大到小。返回的值大于或等于指定的值时条件成立。 string1==string2 string1和string2都为字符的数据，英文字符的大小写将看做不同，这个条件中的等于号必须是2个（绝对相等），条件想等后即执行后面的command EXIST filename 为文件或目录存在的意思。 IF ERRORLEVEL这条语句必须放在某一个命令后面。执行命令后由IF ERRORLEVEL来

Drivers have to pay a fee for driving in busy city streets when there is a great amount of traffic

Drivers have to pay a fee for driving in busy city streets when there is a great amount of traffic. Nowadays, cars are becoming more and more important in the daily lives of the citizens. In the modern society, the car also represents the personal ability. I agree with the statement with three reasons. First of all, the expensive fees will reduce the cars in the city. Then it can improve the quality of the living conditions of the citizens. The control of the key source can decrease the noise and pollution effectively. For example, my uncle lives in Beijing. Five years ago, there were a lot of cars on the road and usually made traffic jams. Then the government begun to collect fees from the cars when there has a traffic jam. Since this policy begun, my uncle has heard less noise than before. Secondly, these fees can become the extra revenue to the government. The huge traffic can create incredible income to the government. These money can do plenty of meaningful things like widened the road and increased the infrastructure. For example, my aunt Judy is a volunteer in the hospital but there often has some problem about funds in the past, then the city begun to collect the fees to solve the problem of the funds.

批处理命令For循环命令详解

分类：2010-06-11 13:0416793人阅读(0) 批处理for命令详解 FOR这条命令基本上都被用来处理文本,但还有其他一些好用的功能! 看看他的基本格式(这里我引用的是批处理中的格式,直接在命令行只需要一个%号) FOR 参数%%变量名IN (相关文件或命令) DO 执行的命令 参数:FOR有4个参数/d /l /r /f 他们的作用我在下面用例子解释 %%变量名:这个变量名可以是小写a-z或者大写A-Z,他们区分大小写,FOR会把每个读取到的值给他; IN:命令的格式,照写就是了; (相关文件或命令) :FOR要把什么东西读取然后赋值给变量,看下面的例子 do:命令的格式,照写就是了! 执行的命令:对每个变量的值要执行什么操作就写在这. 可以在CMD输入for /看系统提供的帮助!对照一下 FOR %%variable IN (set) DO command [command-parameters] %%variable 指定一个单一字母可替换的参数。 (set) 指定一个或一组文件。可以使用通配符。 command 指定对每个文件执行的命令。 command-parameters 为特定命令指定参数或命令行开关。 现在开始讲每个参数的意思 /d 仅为目录 如果Set (也就是我上面写的"相关文件或命令") 包含通配符（* 和），将对与Set 相匹配的每个目 录（而不是指定目录中的文件组）执行指定的Command。 系统帮助的格式:FOR /D %%variable IN (set) DO command 他主要用于目录搜索,不会搜索文件,看这样的例子 @echo off for /d %%i in (*) do @echo %%i pause 把他保存放在C盘根目录执行,就会把C盘目录下的全部目录名字打印出来,而文件名字一个也不显示!

非常齐全用于IGBT驱动和IPM Drivers的光耦型号推荐

非常齐全用于IGBT驱动和IPM Drivers的光耦型号推荐IC输出 高速1Mbps器件有： PS9613 PS8501 PS8502 PS9513 8-pin DIP封装 PS8101、PS9113 PS9213(蠕动5.5mm) 5-pin SOP (SO-5) 封装 PS8802-1/-2 PS8821-1/-2 8-pin Small SOP(SO-8) 封装 超高速10Mbps器件有： PS9617 PS9587 8-pin DIP封装 PS9115、PS9117A、PS9121、PS9214(蠕动5.5mm) 5-pin SOP(SO-5) 封装 PS9817A-1/-2 PS9821-1/-2 8-pin Small SOP(SO-8) 封装 超高速15Mbps (CMOS)器件有： PS9151 5-pin SOP(SO-5) 封装 PS9851-1/-2 8-pin Small SOP(SO-8) 封装 隔离放大器类模拟输出器件有： PS8551 8-pin DIP封装 隔离放大器类数码输出器件有： PS9551 8-pin DIP封装 B、晶体管输出 通用型单级器件有： PS2501series PS2501Aseries PS2513 series PS2561 series 4-pin DIP封装PS2561Aseries PS2561Bseries PS2581 series PS2581Aseries 4-pin DIP封装PS270x series PS2701A series PS2761B series SOP封装 PS2801 series PS2801C series PS2841 series PS2861 series Small SOP封装PS291x series 超小型扁平引脚封装 通用型达林顿器件有： PS25x2 series 4-pin DIP封装 PS2702-1 SOP封装 PS2802-1/-4 Small SOP封装 AC输入单级器件有： PS25x5 series 4-pin DIP封装

BAT(批处理)文件编写详细手册

BAT（批处理）文件编写详细手册 一、批处理文件简介 首先，批处理文件是一个文本文件，这个文件的每一行都是一条DOS命令（大部分时候就好象我们在DOS提示符下执行的命令行一样），你可以使用DOS下的Edit或者Windows的记事本(notepad)等任何文本文件编辑工具创建和修改批处理文件。 其次，批处理文件是一种简单的程序，可以通过条件语句(if)和流程控制语句(goto)来控制命令运行的流程，在批处理中也可以使用循环语句 (for)来循环执行一条命令。当然，批处理文件的编程能力与C语言等编程语句比起来是十分有限的，也是十分不规范的。批处理的程序语句就是一条条的 DOS命令（包括内部命令和外部命令），而批处理的能力主要取决于你所使用的命令。 第三，每个编写好的批处理文件都相当于一个DOS的外部命令，你可以把它所在的目录放到你的DOS搜索路径(path)中来使得它可以在任意位置运行。一个良好的习惯是在硬盘上建立一个bat或者batch 目录（例如C:/BATCH），然后将所有你编写的批处理文件放到该目录中，这样只要在path中设置上c:/batch，你就可以在任意位置运行所有你编写的批处理程序。 简单的说，批处理的作用就是自动的连续执行多条命令。 二、批处理中常用命令： 1.echo命令：表示显示此命令后的字符。字符不需要使用引号。其中，%和^两个字 符不能显示。 2.Echo off命令：表示在此语句后所有运行的命令都不显示命令行本身。即关闭echo 回显功能。可以使用echo on打开。 3.@命令：它是加在每个命令行的最前面，表示运行时不显示这一行的命令行（只能 影响当前行）。 4.call命令：调用另一个批处理文件（如果不用call而直接调用别的批处理文件， 那么执行完那个批处理文件后将无法返回当前文件并执行当前文件的后续命令）。 5.pause命令：运行此句会暂停批处理的执行并在屏幕上显示Press any key to continue...的提示，等待用户按任意键后继续。如果不想显示以上英文提示信息， 可以这样写：pause>null 6.rem命令：表示此命令后的字符为解释行（注释），不执行，只是给自己今后参考 用的（相当于程序中的注释）。 7.::命令：与rem命令等效。 例1：用edit编辑a.bat文件，输入下列内容后存盘为c:/a.bat，执行该批处理 文件后可实现：将根目录中所有文件写入 a.txt中，启动UCDOS，进入WPS等功能。 批处理文件的内容为: 命令注释： @echo off //不显示后续命令行及当前命令行

批处理参数详解

如果没有一定的相关知识恐怕不容易看懂和理解批处理文件，也就更谈不上自己动手编写了 批处理文件是无格式的文本文件，它包含一条或多条命令。它的文件扩展名为 .bat 或 .cmd。在命令提示下键入批处理文件的名称，或者双击该批处理文件，系统就会调用Cmd.exe按照该文件中各个命令出现的顺序来逐个运行它们。使用批处理文件（也被称为批处理程序或脚本），可以简化日常或重复性任务。当然我们的这个版本的主要内容是介绍批处理在入侵中一些实际运用，例如我们后面要提到的用批处理文件来给系统打补丁、批量植入后门程序等。下面就开始我们批处理学习之旅吧。 一.简单批处理内部命令简介 1.Echo 命令 打开回显或关闭请求回显功能，或显示消息。如果没有任何参数，echo 命令将显示当前回显设置。 语法 echo [{ on|off }] [message] Sample：@echo off / echo hello world 在实际应用中我们会把这条命令和重定向符号（也称为管道符号，一般用> >> ^）结合来实现输入一些命令到特定格式的文件中.这将在以后的例子中体现出来。 2.@ 命令 表示不显示@后面的命令，在入侵过程中（例如使用批处理来格式化敌人的硬盘）自然不能让对方看到你使用的命令啦。 Sample：@echo off @echo Now initializing the program,please wait a minite... @format X: /q/u/autoset (format 这个命令是不可以使用/y这个参数的，可喜的是微软留了个autoset这个参数给我们，效果和/y是一样的。) 3.Goto 命令 指定跳转到标签，找到标签后，程序将处理从下一行开始的命令。 语法：goto label （label是参数，指定所要转向的批处理程序中的行。） Sample： if { %1 }=={ } goto noparms if { %2 }=={ } goto noparms（如果这里的if、%1、%2你不明白的话，先跳过去，后面会有详细的解释。）@Rem check parameters if null show usage :noparms echo Usage: monitor.bat ServerIP PortNumber goto end 标签的名字可以随便起，但是最好是有意义的字母啦，字母前加个：用来表示这个字母是标签，goto命令就是根据这个：来寻找下一步跳到到那里。最好有一些说明这样你别人看起来才会理解你的意图啊。 4.Rem 命令 注释命令，在C语言中相当与/*--------*/,它并不会被执行，只是起一个注释的作用，便于别人阅读和你自己日后修改。 Rem Message Sample：@Rem Here is the description.

London Taxi Drivers

London Taxi Drivers 李锡延选注 London taxi drivers know the capital like the back of their hands. Just jump into one of the city's 22,000 distinctive shaped cars and tell the driver your destination. No matter how small and obscure the street is, the driver will be able to get you there without any trouble. The reason London taxi drivers are so efficient is that they have all gone through a very rough training period known as “the knowledge” to get the special licence needed to drive taxis. During this period, which can take from two to four years, the would-be taxi driver has to learn the most direct route to every single road and to every important building in London. To achieve this, most learners go around the city on small motorbikes, practicing how to move to and from different points of the city. Going around London on a small motorbike can have its problems, particularly during the winter. Collin Sinclair, 40, who has been a taxi driver for 15 years, described his training period as a time of blood, s weat and tears. “There was thick snow everywhere and I had to wear my mother's tights because I was so cold,” he said. Learner taxi drivers are tested several times during their training period by government officers. Sinclair thought his exams were a nerve-racking experience. “The officers ask you, …How do you get from Buckingham Palace to the Tower of London?' And you have to take them there in a very direct line. When you get to the Tower, they won't say, …Well done.' They will quickly move on to the ne xt question. After five or six questions, they'll just say, …See you in two months time,' and then you know the exam is over.” Learner drivers are not allowed to work — and earn money — as drivers. Therefore, many of them keep their previous jobs until they obtain their taxi-driving licence. The training period can cost quite a lot, because learners have to pay for their own expenses (getting around London using private transport). The tests they take and a medical exam. Once a new taxi driver has a licence, the next thing he or she has to cope with is the public. Drivers agree that most passengers as Brian Turner, 53, a taxi driver for 30 years, explains: “Your job is to take them where they want to go in a polite and pleasant manner, whatever they are like. After all, if you're unpleasant to your passenger, you won't get a tip.” But sometimes it is not only the tip that is at stake, a taxi driver's job can also be dangerous. Collin Sinclair was once attacked by a passenger who did not