50183 The Afghanistan Investment Climate in 2008 Growth Despite Poor Governance, Weak Factor Markets, and Lack of Innovation THE WORLD BANK FINANCE AND PRIVATE SECTOR DEVELOPMENT SOUTH ASIA REGION THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 GROWTH DESPITE POOR GOVERNANCE, WEAK FACTOR MARKETS, AND LACK OF INNOVATION THE WORLD BANK FINANCE AND PRIVATE SECTOR DEVELOPMENT SOUTH ASIA REGION TABLE OF CONTENTS ACKNOWLEDGMENTS v EXECUTIVE SUMMARY vi INTRODUCTION AND BACKGROUND 1 OVERALL FINDINGS 6 THE GOVERNANCE CHALLENGE 19 GETTING FACTOR MARKETS TO WORK BETTER 27 SERVICED LAND 28 ACCESS TO FINANCE 33 LABOR MARKETS 38 INNOVATIVE AND PRODUCTIVITY-ENHANCING BEHAVIORS 39 CONCLUSIONS 41 ANNEX 1. AFGHANISTAN AT A GLANCE 45 ANNEX 2. THE ENTERPRISE SURVEY 47 SAMPLE 47 METHODOLOGY 49 CHALLENGES AND LIMITATIONS 51 INTERVIEWER PROFILE 52 RESPONDENT PROFILE 52 ANNEX 3. METHODOLOGY FOR PRODUCTIVITY ANALYSIS 53 ANNEX 4. AFGHANISTAN AND COMPARATOR COUNTRIES 56 Currency Equivalents (Exchange Rate effective May 31, 2009) Currency Unit = Afghani (Af) Af 1 = US$0.01907 US$1 = Af 52.45 Government Fiscal Year March 21 ­ March 20 Abbreviations and Acronyms ADB Asian Development Bank AISA Afghanistan Investment Support Agency ARTF Afghanistan Reconstruction Trust Fund DB Doing Business DFID UK Department for International Development FDI Foreign Direct Investment GoA Government of Afghanistan IBRD International Bank for Reconstruction and Development IC Investment Climate ICA Investment Climate Assessment ICS Investment Climate Survey IDA International Development Association IFC International Finance Cooperation IMF International Monetary Fund NBFI Non banking Financial Institution NGO Non Governmental Organization NRVA National Risk and Vulnerability Assessment SME Small and Medium Enterprise USAID United States Agency for International Development Vice President: Isabel M. Guerrero Country Director: Nicholas J. Krafft Country Manager: Mariam J. Sherman Sector Manager: Simon C. Bell Task Team Leader: John F. Speakman iv ACKNOWLEDGMENTS This report was prepared by a World Bank team led by John F. Speakman and comprising Suhail Kassim (lead author and project manager), Niti Bhutani (co-author), Arvind Jain (technical survey advisor) and Nazir Ahmad (survey coordinator). William Byrd, Anne Tully and Birgit Hansl provided helpful guidance and detailed editing comments. Md. Reazul Islam, Jorge Luis Rodriguez Meza, Dhruba Purkayastha, Michael Wong, Richard Nash, Kyoo-Won Oh and Khaleda Atta provided valuable inputs to the preparation of the report. Aza Rashid provided editing support and Parwana Nasiri provided administrative support. Overall guidance was provided by Nicholas Krafft (Country Director, Afghanistan), Ernesto May (Sector Director, South Asia Poverty Reduction, Economic Management, Finance and Private Sector Development Department), Simon Bell (Sector Manager) and Mariam Sherman (Country Manager, Afghanistan). The team also benefited greatly from the advice and guidance of peer reviewers John Nasir, Stephane Guimbert, Joseph Saba, and Douglas Pearce. The firm-level survey was conducted by Altai Consulting and their local partner Noma Consulting under the supervision of Eric Davin, Gemma Stevenson and Muzamel Karim. The Afghanistan Enterprise Survey 2008 was co-financed by the Department for International Development (DFID) and the World Bank. The DFID team was led by Miguel Laric and Chris Bold. The team acknowledges Private Sector and Civil Society Enabling Council, the Ministry of Commerce and Industry (Government of Afghanistan), and the Afghanistan Investment Support Agency (AISA) for their support. Peace Dividend Trust is acknowledged for their support on sharing databases. The survey findings were shared with stakeholders from the Afghan government, private sector, donors, UN and nongovernmental organizations, consultants, and research firms in January 2009, March 2009, and May 2009. There was broad consensus about the key findings. The inputs received through these interactions are gratefully acknowledged. v EXECUTIVE SUMMARY The key finding of the survey is that strong private sector growth, albeit from a small base, is taking place in Afghanistan despite poor governance, weak factor markets, and a lack of innovation. 1. This report summarizes the findings of an enterprise survey con- ducted from September to November 2008, covering 1,066 firms in 10 Afghan cities. It complements an earlier survey of 338 firms in 2005. Both surveys used the standard World Bank enterprise survey methodology. This report presents the results of the 2008 survey, along with some comparisons with the 2005 survey and some international comparisons. The survey provides a good representation of firms operating in the major urban centers, but it does not cover rural firms, firms that tried to enter the market but failed, or firms that were oper- ating but exited. 2. This work is part of a broader effort by the World Bank to under- stand growth and the prospects for growth in Afghanistan, and to stimulate dialogue on how to develop the levels of growth necessary to secure livelihoods and economic independence for the Afghan people. The report does not provide much in the way of recommendations; rather, it identifies the issues, summarizes the reform status, and, where action is required, identifies the next steps. GROWTH FROM A SMALL BASE 3. The survey begins by confirming many of the key characteristics of the private sector in Afghanistan. It is small, and it includes few foreign firms (less than 2% of the sample), few exporters (7% of the sample), and few firms owned by women (3% of the sample). In the past two years, the rate of new firm entry appears to be significantly lower than immediately after the fall of the Taliban in 2002. Nevertheless, the pri- vate sector is growing fast: national accounts data suggests that the industrial and service sector has doubled in size since 2005. Average revenue growth in the surveyed firms is very high--220 percent-- because of some strong performers. The survey suggests that this rate of growth will continue, as 77 percent of the surveyed firms plan to expand in the near future. vi THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 KEY FINDINGS 4. Despite the growth, the survey identifies a number of significant con- straints clustered around the issues of poor governance, weak factor markets, and lack of firm-level innovation. While the survey cannot explain this contradiction between growth and these weaknesses, it is possible that strong demand arising from increased aid flows and catch- up effects as the economy moves from command to market mode are contributing factors. The survey results suggest that firms are quietly resolving and managing the bewildering array of constraints they face. The question is, what would growth be like without these constraints? 5. Six constraints emerge as dominant in all our analyses--by sector, by firm size, and by region. These constraints are (1) weak policy enforcement; (2) poor provision of electricity; (3) crime, theft and dis- order (4) corruption; (5) access to land; and (6) access to finance. These constraints are more or less the same as those found in the 2005 survey, with a marked deterioration in policy enforcement and secu- rity related to crime, theft, and disorder. While the constraints are broadly common, there are some interesting variations. For example, the single biggest constraint is different in each city. Governance issues dominate in the less secure areas of the country, while factor markets are more constraining in the more secure areas. Women-owned firms identify the constraints as worse in all cases. The survey proposes addi- tional work to understand the underlying causes. POOR GOVERNANCE 6. The governance challenge has five main components: (1) policy enforcement; (2) crime, theft, and disorder; (3) corruption; (4) the court system; and (5) business licensing and permits. Perceptions have become more negative on all six components since 2005, except for corruption, which is perceived as near 2005 levels. The less affected firms below the radar screen are small and located in stable areas. 7. Policy enforcement is a dominating issue in the survey. It is hard to define accurately, but it has important productivity consequences, as firms that are constrained by it are 32 percent less productive than firms that are not. We have defined it as "state ineffectiveness in designing and implementing consistent policies." Only 14 percent of the surveyed firms vii THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 reported consistency and predictability in the interpretation of laws. This point is critical, because it concerns an issue that firms cannot man- age. They cannot control for unexpected changes in the rules of the game. While more work is required to fully understand the dimensions of this issue, it is clearly a major red flag for policymakers. 8. Crime, theft, and disorder have increased dramatically since the 2005 survey, with one in five firms reporting losses resulting from theft, rob- bery, arson, or vandalism. These firms are, on average, 63 percent less productive than firms that did not suffer such losses. Larger firms, retail- ers, exporters, and firms in Herat (where 90% of firms identified crime as a major constraint) are most affected. Firms have responded by pay- ing for security and making protection payments. 9. The World Bank's 2009 Doing Business report shows that Afghanistan's business regulatory environment is quite weak (ranked 162nd out of 181 economies). This indicator is very important for entrepreneurs who are trying to enter the market; however, the sur- vey did not cover this group. For this reason, this indicator emerges as one of the least constraining factors in the governance group, with just 23 percent of the sample seeing it as an issue. This can further be explained in part by the age of surveyed firms: most firms are well established by now and no longer have to face many of the regulatory burdens identified in this indicator. 10. In two instances, quantitative indicators identify a governance problem that is not confirmed by perception data. In the case of the courts, 84 percent of respondents say the court system is corrupt and unfair, while 19 percent consider this a major constraint to doing busi- ness. Corruption was rated as a problem by 44 percent of the respon- dents, and the survey shows that it touches virtually every aspect of a firm's dealing with government. Firms that are not able to manage cor- ruption are 26 percent less productive. WEAK FACTOR MARKETS 11. Little apparent progress has been made on the key factor markets of serviced industrial land, capital, and labor since the last survey. This is unfortunate, as these factors directly affect firm productivity and firm viii THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 entry into the market. For firms that want to grow, these issues are partic- ularly constraining. When asked what their single biggest constraint was, access to formal finance was reported by 22 percent of the surveyed firms and access to land by 13 percent. In Kabul, 83 percent of firms that tried to acquire additional land in the past three years failed to do so. 12. Electric power is a hugely constraining factor, both on its own as a direct "tax" on productivity and combined with land as a fundamental barrier to entry. Firms report losing, on average, 9 percent of output because of power losses. Firms that have access to the grid are 49 percent more productive than those that do not. Generators provide 77 percent of electricity, which affects profitability. Other infrastructure sectors-- such as telecommunications, transport, and water have also been identi- fied as lower level constraints. 13. Modest improvement was reported in access to formal finance, but it remains a major constraining factor. Only a very small propor- tion of firms (5% of the sample) borrow from the formal financial sec- tor, while 85 percent of all new investments are financed out of retained profits. Among the firms that do not seek formal credit are the 26 percent that report seeking Islamic financing. One area of improve- ment is the proportion of firms that have a bank account: 51 percent in 2008 compared with 30 percent in 2005. 14. Few firms identify lack of skills as a problem. Only 18 percent of respondent firms say they cannot get the skills they need, and only 20 percent identify this as a major constraint. Nevertheless, the lack of skilled workers is more of an issue now than it was in 2005, when only 3 percent of firms identified it as a major problem. This trend should be of concern to policymakers. LACK OF INNOVATION 15. A number of indicators identified lack of innovation as an emerg- ing area for policymakers: 94 percent of surveyed firms do not use foreign licensed technology; 86 percent do not offer on-the-job train- ing; 95 percent do not have internationally recognized quality certifi- cation; and 70 percent do not have internet connectivity. Research shows that firms that engage in innovative practices are significantly ix THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 more productive and competitive. It is not surprising to find very few exporting firms in the sample. CONCLUSIONS 16. The survey findings generally validate the government's current reform program, including efforts to improve governance and strengthen factor markets. Clearly, more efforts are needed in these areas, as the survey reports little measurable progress since 2005. Par- ticular emphasis is required in the area of policy enforcement--it is crucial to ensure consistency in policy design and implementation to attract investment into Afghanistan. 17. The report also identifies a group of emerging issues that include the interlinked issues of competitiveness, innovation, and diversifica- tion. Despite strong growth, policymakers should be concerned about the lack of entry of new firms--especially foreign firms--and the lack of innovative behavior. These two factors indicate a lack of competi- tiveness and warrant further research. An undiversified manufacturing sector that is overwhelmingly (approximately 95%) linked to the agri- culture and agro-processing sectors is also of concern. 18. The strong growth trend itself needs more evaluation. A country that has weak governance, poor factor markets, and firms that are not innovative is unlikely to achieve sustainable long-term growth. 19. The report discusses the issues noted above, summarizes the sta- tus of reforms, and suggests some next steps, including further analytic work on a number of topics, public private dialogue on certain issues, and stronger government reform efforts. x THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 INTRODUCTION AND BACKGROUND 1.1 This report summarizes the findings of an enterprise survey that took place from September to November 2008. The report iden- tifies some key issues, summarizes the reform status, and suggests next steps where action is required.1 After the report is discussed with policymakers, a series of research questions will be identified that will be the subject of additional papers by the World Bank and its international partners. For example, a detailed financial sector assessment is planned to help answer questions about access to finance. After the additional reports and/or papers are completed, the World Bank will undertake a comprehensive economic review. This survey is expected to be an important input into all these research studies. 1.2 This survey report will help the government of Afghanistan think through its approach to private sector development. Histori- cally, there has been a dearth of information and reliable statistics about Afghanistan's economy. This report reviews the constraints that firms currently operating in Afghanistan face and provides a basis for possible policy recommendations to address these con- straints. It is hoped that the report will be a useful tool to support investment climate reforms and enhance the private sector dialogue in Afghanistan. 1.3 The 2005 Investment Climate Assessment (ICA) is an important point of comparison. The World Bank prepared an Afghanistan ICA in December 2005; it was one of the first post-conflict ICAs and used enterprise survey data on 338 firms and other secondary sources to assess Afghanistan's investment climate.2 The 2008 survey is a follow- up to the ICA, in keeping with the objectives of monitoring the progress in removing investment climate constraints, sustaining dia- logue with the various stakeholders to stimulate policy reform, and 1 It is important to distinguish this report from a full investment climate assessment (ICA), which is comprehensive and contains detailed economic perspectives and policy recommendations. 2 The Investment Climate in Afghanistan: Exploiting Opportunities in an Uncertain Environment (2005).World Bank. 1 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 providing input for the reform agenda at the central and provincial lev- els. A key objective of this report is to highlight areas of progress or lack of progress since the previous survey and to compile data to inform donors. It uses an evidence-based approach to foster dialogue and help develop a monitoring and evaluation framework. 1.4 An enterprise survey of 1,066 firms was conducted across 10 provinces in 2008. The 2008 survey generates firm-level quantitative and qualitative information through face-to-face interviews with sen- ior managers.3 These data can be used to evaluate firm-level perform- ance, identify constraints to doing business, and prioritize reform areas. The survey indicators are also useful in comparing Afghanistan's investment climate with those in other countries and for assessing the investment climate of regions within Afghanistan. 4 1.5 Firms in the survey are examined over five dimensions: city, sec- tor, size, type, and market (figure 1). 1.6 The report undertakes a limited analysis across time and space. Across time, the survey incorporates an analysis of 58 panel firms. This relatively small number reflects high sample attrition (over 80%) among the firms surveyed in 2005. There are various possible reasons for the attrition, including firm closure, difficulty in locating firms, and survey fatigue. Across space, the report undertakes an analysis across 10 Afghan cities and six comparator countries. 3 Details of the survey methodology and its limitations are provided in annex 2. 4 Note that the numbers presented in this report may differ from the indicator values as pre- sented on the Enterprise Surveys website (http://www.enterprisesurveys.org/). The reasons for this are threefold: (1) For the purposes of this report, unregistered firms are included in the analytic sam- ple whereas Enterprise Surveys methodology focuses on "formal" firms where formality is defined as being registered; (2) To gain a more robust snapshot of the Afghan non-agricultural economy, the Afghanistan Enterprise Survey included a strata of firms whose business activity (e.g. educational training, health services) are typically not interviewed in a standard World Bank Enterprise Survey. For this report, those firms are included in the analytic sample whereas for comparability reasons, firms whose business activity typically fall outside an Enterprise Survey are excluded in the computa- tion of the website indicators; (3) In the Enterprise Surveys website, outliers, values that are 3 or more standard deviations from the mean are removed for website indicator computation purposes for select indicators. Further documentation regarding removal of outliers can be found on the Enter- prise Survey website. 2 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 1 Survey Profile Coverage by Province Mazar-e- Sharif (13%) Kunduz Pule-Khumri Bamyan Kabul Jalalabad Herat (29%) (12%) (12%) Ghazni Khost Kandahar (11% ) Coverage by Sector Coverage by Firm Size 74 (7%) 65 (6%) 225 (21%) 201 (19%) 399 (37%) 231 (22%) 726 (68%) 211 (20%) Other services Retail Large (100 or more) Small (5 to 19) Construction Manufacturing Medium (20 to 99) Micro (<5) Coverage by Firm Type Exporter and Non-Exporter Firms 72 (7%) 419 (39%) 647 (61%) 981 (93%) Unregistered Registered Non-exporter Exporter 3 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 BOX 1 A User's Guide to Data in the Enterprise Survey The assessment of Afghanistan's investment climate in the report is based on five kinds of data. 1. Perception data. This information was collected as part of the enter- prise survey. Two kinds of perception data are covered in the survey: (1) single biggest obstacle, a rank indicator derived from the respon- dent's choice from a list of 15 obstacles in the business environment, and (2) severe constraint, a rating indicator. Respondents were asked whether they considered a given constraint to be not an obstacle, a minor obstacle, a moderate obstacle, a major obstacle, or a very severe obstacle. This report focuses on constraints rated "major" or "very severe." 2. Quantitative data. These data were also collected as part of the survey. The questions in this category are designed to gather spe- cific, objective information about a given issue rather than infor- mation based on perception. Perception data and quantitative data, although collected as part of the same survey, may not yield the same interpretation. For example, Afghan firms' perceptions relating to corruption are not as negative as one might expect, even though quantitative data indicate that corruption is a serious con- straint to productivity and growth. Placing too much importance on perception data is not recommended. These data are useful in pointing toward trends, but they can be misleading if respondents do not have adequate knowledge to accurately rate a constraint or if they underrate it because they have become accustomed to it over time.5 3. Data across time (perception and quantitative). 58 of the 338 firms surveyed in 2005 also participated in the 2008 survey. However, given the small number of panel firms, only a limited panel analysis has been undertaken in this report. 4. Data across space (perception and quantitative). Two sets of spa- tial data are covered in the report: (1) across cities--this survey covers 10 Afghan cities--and (2) across countries. The source of (continued) 5 Perceptions data that measure the relative severity of issues do not lend themselves to analyzing changes over time. For example, at a certain point, crime might be the most prominent concern, and people might say that it is more of a concern than, say, transport issues. A year later, crime levels might be the same but transport problems might have increased dramatically, so survey respondents might say that transport is a bigger concern than crime. An incorrect interpretation of these data would be that crime had decreased. The truth is that crime stayed the same but transport problems increased. 4 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 these data is the World Bank's enterprise survey database. Num- bers are comparable across countries, because they are based on questions from a common core survey. We chose six countries as a benchmark against which to measure the performance of Afghan firms. Comparator countries have geographic, socio-cultural, or economic similarities with Afghanistan and have been surveyed in the recent past. Four of the comparator countries were Ethiopia, Pakistan, Tajikistan, and Yemen. We also included Thailand and the Democratic Republic of the Congo--respectively, the highest and lowest ranked emerging market economies in the World Bank's Doing Business 2009 report. Other countries are occasion- ally mentioned to emphasize a point. 5. External data. These data are used to support the hypotheses and findings of the report. Some of our sources are Afghanistan's Central Statistics Office, www.icasualties.org, the World Economic Forum, and the Development Economics Longitudinal Business Database. 1.7 This report provides information that can help answer some key questions of policymakers. However, while the report provides information on a large segment of the Afghan private sector, the report does not capture information from some firms: I Firms that tried to enter the market but could not. I Firms that entered the market but failed and dropped out. I Firms in rural areas. 1.8 The report flags issues but does not offer detailed recommenda- tions. The goal is to provide information and identify issues for policy- makers. It is the responsibility of the Government of Afghanistan and other stakeholders to act on the findings of the report. 5 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 OVERALL FINDINGS 2.1 The key finding is that enterprises are showing surprising levels of dynamism in the face of a poor and deteriorating governance environ- ment, continuing severe problems with factor inputs, and a lack of innovation by firms. 2.2 The private sector in Afghanistan is in a formative stage, with the following characteristics: I The private sector is dominated by agribusiness, construction, and trading firms (figure 2). Private industry's share of the economy is quite small, with National Risk and Vulnerability Assessment (NRVA) data suggesting that only a little over 5 percent of house- holds earn income from industrial activity.6 I There are few registered firms--the Afghanistan Investment Sup- port Agency (AISA) database listed just over 5,000 firms in the ten surveyed cities. Furthermore, there are very few large firms--we were able to identify only 74 firms with more than 100 employees. FIGURE 2 Sectoral Profile of GDP in Afghanistan Sectoral Share of Private Sector GDP 28% 35% 2% 16% 19% Agribusiness Non agribusiness industry Construction Trading Other services 6 National Risk and Vulnerability Assessment 2005, Afghanistan Central Statistics Office. 6 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 I Afghan firms primarily produce for the domestic market. In our 2008 sample, less than 1 percent of firms reported selling their prod- ucts mainly in international markets; less than 1 percent of the sales were direct exports, while indirect exports accounted for 5 percent.7 The survey also found that retail firms have a substantial presence in terms of indirect exports: 21 percent of retail firms reported such sales compared with 2 percent of manufacturing firms and 0.04 per- cent of construction firms. I Less than 2 percent of the firms surveyed in 2008 were foreign owned, which is an indicator of the low levels of foreign direct investment (FDI) flows into Afghanistan. I Afghan firms in general do not engage in innovative or productivity- enhancing behaviors such as quality certification or technology licensing. 2.3 The survey shows that Afghan businesses have been growing and want to grow more. Afghan businesses are ambitious in their expansion plans: 77 percent of respondents plan to expand in 2009. Between 2005 and 2008, the average growth in sales for the panel firms was 220 percent and median growth was 92 percent, with two- thirds of panel firms recording consistent growth. This is further cor- roborated by national accounts data, which report a doubling of industrial and service sector GDP since 2005 (figure 3). The strong FIGURE 3 Growth in Private Sector GDP Private Sector GDP 150000 100000 AFs 000's 50000 0 2005 2006 2007 2008E Industry Services Source: Afghanistan Central Statistics Office. 7 An example of indirect exports here is re-exports. 7 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 firms are getting stronger: the intention to expand is greater for large and medium-sized firms than for micro and small firms, and for reg- istered firms than for unregistered firms. This is particularly true for Kabul-based firms. Eight percent of all firms say they plan to expand internationally. 2.4 Analyzing the single biggest obstacle identified by firms supports the notion of latent dynamism in the private sector. When firms were asked to pick just one constraint that is the single biggest obstacle to their business, factor markets led the way. Access to finance was iden- tified as the single biggest problem for 22 percent of the sample, fol- lowed closely by electricity (21%), and access to land (13%). The gov- ernance-related indicators--crime, theft, and disorder (17%), policy enforcement (15%), and corruption (6%)--were cited as the single biggest obstacle less often than finance or electricity. 2.5 However, when firms were asked to rate the severity of individual constraints, both governance and factor markets emerged as major issues (figure 4). The following are the top six constraints: I Policy enforcement I Electricity I Crime, theft, and disorder I Corruption I Access to land I Access to finance 2.6 The main constraints are consistent across the survey.8 These constraints rank higher than others not only for Afghanistan as a whole but across all sectors, across all firm sizes except small firms, across most cities, and irrespective of firm type or market.9 However, 8 One constraint ranks higher in severity than another constraint if the percentage of firms that cite it as a major or very severe obstacle is higher than the percentage of firms that cite the other con- straint as major or very severe. 9 Ninety-four percent of the surveyed firms reported one of these six constraints as their single biggest obstacle. 8 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 4 Major Business Problems Percentage of Firms Citing Factor as a Major or a very Severe Obstacle Policy Enforcement 70 Electricity 66 Crime, theft, and disorder 60 Corruption 44 Access to Land 41 Access to Finance 38 Telecommunications 29 Tax Rates 28 Transport 24 Business Licensing and Permits 23 Practices of Comp. in Informal Sector 22 Inadequately Educated Workforce 20 Courts 19 Customs & Trade Regulations 15 Tax Administration 15 Labor Regulations 4 0 10 20 30 40 50 60 70 80 Percentage TABLE 1 Ranking of the top six investment climate constraints by sector Policy Crime, Theft, Access to Access to Enforcement Electricity & Disorder Finance Corruption Land Manufacturing 1 2 3 4 5 6 Construction 1 4 3 6 2 5 Retail 1 3 2 5 4 6 TABLE 2 Ranking of the top six investment climate constraints by firm size Policy Crime, Theft, Access to Access to Enforcement Electricity & Disorder Finance Corruption Land Micro 1 3 2 4 6 5 Small 2 3 4 7 6 5 Medium 1 2 3 6 4 5 Large 1 3 2 5 4 6 9 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 TABLE 3 Ranking of the top six investment climate constraints by city Policy Crime, Theft, Access to Access to Enforcement Electricity & Disorder Finance Corruption Land Kabul 1 2 3 5 4 6 Herat 1 5 2 4 3 7 Kandahar 1 2 3 13 4 6 Mazar-e-Sharif 1 4 2 3 7 8 Jalalabad 3 1 2 10 4 11 TABLE 4 Ranking of severe constraints by firm type Policy Crime, Theft, Access to Access to Enforcement Electricity & Disorder Finance Corruption Land Registered 1 2 3 6 4 5 Unregistered 1 2 3 5 6 4 TABLE 5 Ranking of severe constraints by market Policy Crime, Theft, Access to Access to Enforcement Electricity & Disorder Finance Corruption Land Exporter 1 3 2 6 4 5 Non-exporter 1 2 3 6 4 5 the ordering of the top six constraints varies within each group with policy enforcement almost always emerging as the top constraint. We have defined the most severe constraint--policy enforcement-- as meaning "state ineffectiveness in designing and implementing consistent policies." Comparing these findings with a recent Depart- ment for International Development (DFID) study on binding con- straints is illuminating. The DFID report highlighted the failure of governance: "From a public policy perspective, the most relevant result is confirmation that in Afghanistan, private investment deci- sions are being negatively affected by the uncertain environment, absence of rule of law, unpredictable tax structures and by widespread 10 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 corruption."10 While our survey confirms these findings, it also points to weak factor inputs and lack of innovation by firms as acute shortcomings in the country. 2.7 Since 2005, there has been a serious decline in the perception of policy enforcement and crime prevention, and not much improve- ment elsewhere (figure 5). This is consistent with security statistics, which include coalition military deaths as one measure (figure 6). Even though perception data indicate that some issues had become less of a problem, the quantitative measures show little improvement in any category. For example, with respect to land, the Doing Business results showed no reduction between 2008 and 2005 in the 252 days it took to register property in Kabul, yet the enterprise survey recorded improved perceptions. In the area of corruption, there was a 20 percent increase in the informal payment required to secure a contract, yet sur- veyed perceptions improved. This could reflect the relative importance of other issues, as well as the maturation of businesses, many of whom started in the post-Taliban period (figure 7). The older firms have resolved--or, more likely, learned to live with--these constraints. FIGURE 5 Major Business Problems (Panel) Percentage of Firms Citing Constraint as Major or very Severe (Panel) Policy Enforcement 15 60 Electricity 59 59 22 Crime, Theft, and Disorder 59 Tax Rates 50 43 Access to Land 63 40 Access to Finance 50 40 Corruption 67 39 28 Telecommunications 33 24 Transport 33 Business Licensing and Permits 9 32 Customs & Trade Regulations 41 29 Practices of Comp. in Informal Sector 32 27 Courts 12 24 3 Inadequately Educated Workforce 20 Tax Administration 39 15 Labor Regulations 2 5 0 10 20 30 40 50 60 70 Percentage Panel firms 2008 Panel firms 2005 10 Understanding Afghanistan Report, DFID, 2008. 11 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 6 Coalition Fatalities Coalition Fatalities, Afghanistan (Oct 2001 ­ Feb 2009) 350 300 250 Fatalities 200 150 100 50 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 Month Source: HSRP; iCasualties.org. FIGURE 7 Ages of Surveyed Firms 600 500 400 300 200 100 0 Less than 3 yrs 3 to 7 yrs 8 to 10 yrs 10 and above yrs 2.8 The aging profile of surveyed firms is unusual. Figure 7 shows that very few surveyed firms entered the market in the past two years. While the survey does not explain this phenomenon, it is relatively unexpected for a country that is moving from a command economy toward a market economy. 2.9 A review of firms constrained neither in terms of policy enforce- ment nor corruption identifies "below-the-radar" firms (figure 8). Such firms (one in every four surveyed), we believe, should enjoy strong productivity benefits relative to their peers. These are typically smaller firms located in relatively secure and politically stable areas. According to our analysis, firms that are constrained by weak policy enforcement are 32 percent less productive than firms that are not, and those con- strained by corruption are 26 percent less productive. For policymakers, the analysis suggests the need to support larger firms, which in most economies are the engines of productivity and innovation. The security 12 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 8 Firms Not Constrained by Policy Enforcement or Corruption Percentage of Firms not Constrained by Both Policy Enforcement and Corruption 50 43 45 Percentage 40 33 33 32 35 28 30 25 22 24 24 25 25 18 20 20 15 17 15 9 10 10 3 5 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Mazar-e- Sharif Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Non-Exporter Overall By Sector By Province By Size By Type By Market findings are also sobering and illustrate the benefit the private sector will gain from improved security. 2.10 While the main constraints are consistent across the five main cities, some interesting differences emerge. In general, the same top six constraints (policy enforcement, electricity, crime and theft, corrup- tion, land, and finance) are reported as the biggest problems in all five cities, although the order varies (table 3; figure 9). And when firms identified their single biggest obstacle, strong differences emerged among the cities (figure 10). Some of these differences can be explained; for example, electricity is a major problem in Jalalabad because it is not connected to the national transmission grid, and access to finance in Kabul because it is the most economically dynamic city and there is a greater unmet need for finance. FIGURE 9 Main Constraints Faced by Firms in Each City Key Constraints Faced by Firms in each City 120 100 96 89 85 88 89 76 76 80 59 62 63 61 61 53 52 53 57 60 48 52 46 48 50 40 33 34 27 22 20 23 23 21 20 5 0 Policy Electricity Crime, Theft, Corruption Access to Access to Enforcement and Disorder Land Finance Kabul Herat Kandahar Mazar-e-Sharif Jalalabad 13 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 10 Single Biggest Obstacle in Each City Electricity Access to (75%) land (39%) Access to finance (32%) Policy enforcement (33%) Corruption (23%) 2.11 While Mazar-e-Sharif reports better perceptions compared with the national average in 14 of 16 constraints, the quantitative indica- tors reveal a mixed picture. The difference is most marked for electric- ity, which just one in three Mazar-e-Sharif firms considers a severe constraint, compared with two out of three firms in the overall survey. Other areas in which Mazar-e-Sharif fares significantly better than the national average are corruption, access to land, and transport. How- ever, while the quantitative governance indicators for Mazar-e-Sharif are strong and support the perception findings, this is not true for access to finance and access to land. For instance, of the 142 surveyed firms in Mazar-e-Sharif, only 4 applied for additional land and 1 applied for a construction permit. The survey also reveals that the largest percentage of firms in Mazar-e-Sharif (39%) reported land as the single biggest obstacle. This, combined with the finding that Mazar-e-Sharif reports the lowest percentage of firms that plan to expand in the near future, raises the question as to why businesses are not growing or showing the dynamism they should in Mazar-e-Sharif (given its strong perception and governance indicators). It could be hypothesized that Mazar-e-Sharif 's obvious strengths on governance have resulted in the perception that it has a relatively good business environment, but its weakness in factor inputs may be causing disin- terest in the private sector in further investing and expanding in Mazar-e-Sharif. One clear finding of the survey is that different cities have different investment climates. 14 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 11 Number of Firms that have One or More Women as Owners Number of Firms that have at Least One Female Owner 40 35 35 29 31 30 25 21 20 15 9 8 9 11 10 7 5 6 4 1 1 1 2 3 5 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporters Non- Exporters Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 2.12 Registered firms11 and exporters12 perceive themselves as more con- strained in several ways. Registered firms are more constrained than unregistered firms in perceptions of all factors except access to finance and courts. Exporters find themselves more constrained than non-exporters with respect to crime, corruption, customs and trade regulations, an inad- equately educated workforce, transport, and telecommunications. 2.13 The number of firms with at least one woman as the owner has increased since 2005 from 6 firms (2% of the sample) to 35 firms (3% of the sample) (figure 11). Such firms are mostly small, registered, and non-exporting firms. 2.14 Women are participating in the Afghan private sector, but not in large numbers. Women constituted 1 percent of top managers, 3 percent of permanent full-time production employees, and about 2 percent of permanent full-time nonproduction employees in the surveyed firms. Of the 15 top women managers, 12 were from the professional, technical, or scientific sector, including teaching and health professionals. The 9 per- cent of manufacturers that employ women are from a more eclectic mix of business activities, spanning production of biscuits, rugs, cabinets, handicrafts and electrical appliances, amongst others (figure 12). 11 In this survey, a "registered firm" is a business that is registered with a central government body; for example, the Afghanistan Investment Support Agency (AISA) or any government ministry. An "unregistered firm" is a business that is not registered with any organization or only registered with the local municipality, business association, or union. The 2008 Enterprise Survey included 419 unregistered firms (39% of the surveyed sample). There could be a correlation between registered and formal firms, and between unregistered and informal firms; however, formality cannot be assumed on the basis of registration status. 12 Firms are classified as "exporters" if they report a higher proportion of sales that are direct or indirect exports, and as "non-exporter firms" if all their sales are national sales. The 2008 sample includes 72 exporter firms and 981 non-exporter firms. 15 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 12 Roles of Women in the Surveyed Firms 25% 20% 20% 15% 10% 9% 5% 3% 1% 0% Owners of this Top Manager Employees in the Employees in all Establishment Manufacturing Other Sectors Sector 2.15 Women-owned firms, where there is at least one female owner, are more constrained as compared to the typical Afghan firm. Percep- tions of constraint severity in such firms are higher for 15 of the 16 constraints; although the top 5 constraints are the same as those for the overall sample (figure 13). Access to finance was ranked as the single biggest obstacle by the majority of these women-owned firms. 2.16 A comparison of Afghanistan with comparator countries reveals similarities in the major constraints and the degree of dissatisfaction. Table 6 shows that emerging market economies such as Thailand can be a source of best practices. (See annex 4 for more comparisons.) Afghanistan can learn much from other developing countries. 2.17 A simple productivity analysis of manufacturing firms reveals the extent to which specific factors can constrain performance. Effective FIGURE 13 Constraints Faced by Firms with at least One Female Owner Percentage of Firms Citing Factor as a Major or Very Severe Obstacle (All, Female-owned) Policy Enforcement 70 74 Electricity 66 69 Crime, Theft, and Disorder 60 63 Corruption 44 58 Access to Land 41 53 Inadequately Educated Workforce 20 51 Access to Finance 38 48 Courts 19 44 Business Licensing and Permits 23 39 Telecommunications 29 38 Transport 24 38 Tax Rates 28 33 Tax Administration 15 30 Practices of Comp. in Informal Sector 22 21 Customs & Trade Regulations 15 17 Labor Regulations 4 14 0 10 20 30 40 50 60 70 80 Percentage Female owned firms All firms 16 TABLE 6 A Snapshot of Quantitative Indicators Across Comparator Countries* % contract value paid % senior % firms with % firms using as informal % total management internationally technology Doing Business Duration of Annual cost payment annual sales time spent on % new recognized licensed from % firms 2009 Rankings power outages of security to secure to get things government investment quality foreign that have (181 countries) (hours) (% of sales) contract done regulations from banks certification companies a website Afghanistan 162 11 3 6 2 7 2 5 6 15 Yemen 98 3 - 16 9 16 9 19 17 - Pakistan 77 2 4 1 34 2 - 9 3 95 Ethiopia 116 - 1 - - 4 11 4 4 18 Tajikistan 159 - 3 - - 12 21 17 25 21 Thailand 13 - - - - 1 75 45 - 34 DR Congo 181 18 3 81 84 6 3 4 4 6 Note: Yellow boxes indicate best practices, while red boxes indicate worst practices. 17 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 policies, a reduction in corruption levels, a secure environment, power, technology, and modern management practices are all critical factors that drive productivity. Details of the methodology are presented in annex 3. The key findings are show in box 2. BOX 2 Key Findings From the Productivity Analysis · Firms that have access to the grid are 49 percent more productive than firms that do not. Not having electricity negatively and signifi- cantly affects productivity. · Corruption and weak policy enforcement negatively affect produc- tivity. A negative correlation can be established between productiv- ity and perceptions relating to corruption and weak policy enforce- ment. Firms that perceive policy enforcement as a major or very severe constraint are 32 percent less productive, while firms con- strained by corruption are 26 percent less productive. However, the perception estimates are only moderately significant. · Firms that reported losses from theft, robbery, arson, or vandalism are less productive than firms that have not reported such losses. This result is invariant to different regression specifications and is statisti- cally significant in all results. In our final specification, a firm reporting output losses due to crime, theft, and disorder was found to be 63 per- cent less productive than a firm that did not suffer such losses. · Younger firms and firms that have an import license have higher productivity. Firms that have an import license1 are 88 percent more productive than firms that do not. Younger firms were found to be more dynamic and productive than older firms. · Mazar-e-Sharif emerges as the most productive city. The productivity gap between firms located in Mazar-e-Sharif and those in other cities covered by the survey was 170 percent and was statistically significant. · Unregistered firms and micro firms are only slightly less productive. Registered firms were found to be just 2 percent more productive than unregistered firms, but this estimate is not statistically significant. Small, medium, and large firms fare better than micro firms with respect to productivity; this estimate is also not statistically significant. There could be a problem of measurement errors in this analysis due to limited sam- ple size and possible data quality issues; also, there are econometric issues that cannot be resolved without panel data. We have nonetheless attempted a limited productivity analysis within the existing constraints, with the caveat that the findings should be viewed strictly as indicative. However, given that most of the productivity analysis findings triangulate well with the perception survey results, as well as with other stud- ies on Afghanistan, there is a certain degree of confidence in the broad messages emerging from the productivity analysis. 1 In Afghanistan, a Trade License is issued to allow businesses to engage in both import and export activities. 18 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 TABLE 7 Firms Face Huge Productivity Taxes Problem Area Indicator SALES Output losses due to power outages 9% of sales Output losses from theft, robbery, arson, or vandalism 11% of sales COST OF SALES AND OVERHEADS Cost of security 3% of sales Percentage of contract value paid to secure contract 6% of sales Bribes to get things done 2% of sales 2.18 The impact of productivity taxes on firms is huge. The survey identified a number of productivity taxes that directly affect firms' bot- tom lines (table 7). There could be other productivity taxes that were not captured by the survey. 2.19 Several strong positive findings bode well for the future. While significant challenges remain to be overcome, the survey identifies some good news for entrepreneurs and investors--see box 3. THE GOVERNANCE CHALLENGE 2.20 The survey highlights governance13 issues of policy enforce- ment, corruption and crime, and security as major problem areas. As the survey shows, these factors impose a tax on productivity and competitiveness. Poor governance also limits the predictability of government interactions. The fundamental connection between gov- ernance and development is illustrated in figure 14. This finding is 13 "Governance" refers to the traditions and institutions by which authority is exercised in a coun- try. This includes the process by which governments are selected, monitored, and replaced; the capacity of the government to effectively formulate and implement sound policies; and the respect of citizens and the state for the institutions that govern economic and social interactions among them (http://info.worldbank.org/governance/wgi/index.asp). 19 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 BOX 3 Celebrating the Good News in the Survey · Private sector exhibits latent dynamism. There is clear evidence of latent growth and investment potential in Afghanistan. Seventy- seven percent of firms plan to expand in 2009, and 8 percent plan to expand internationally. Two out of three panel firms recorded growth, with average sales growth at 220 percent. Growth was higher among large and medium-sized firms, registered firms, and firms based in Kabul. · Overall findings are positive. Panel firms recorded improved per- ceptions on several key indicators since 2005, including corrup- tion, skills, access to finance, access to land, transport, tax rates and administration, and customs and trade regulations. Moreover, some Doing Business indicators including starting a business, employing workers and paying taxes ­ reinforce the survey find- ings that business licensing and permits, lack of skills, tax rates and tax administration are not perceived by firms as severe con- straints. Afghanistan also performs better than comparator coun- tries on several quantitative indicators. Mazar-e-Sharif reports better perceptions compared with the national average in 14 of 16 constraints, and the productivity gap between firms in Mazar-e- Sharif and those in other surveyed cities is 170 percent. The num- ber of firms where at least one owner is female has increased since 2005. · Governance displays elements of hope. The survey showed that in areas where security and services are better, productivity improves dramatically. · Factor markets lead the way. Rigidities in labor markets and availability of skills are not generally perceived to be a problem. The majority of workers in Afghan firms is literate. Although only half of all firms have a bank account, this is an improvement over 2005. Several indicators of physical infrastructure--including transport, telecommunications, and water--were not found to be big issues. · Innovative and productivity-enhancing behaviors exist in some Afghan firms. The survey established that firms that use technology and modern management practices, and those that trade interna- tionally, do better. About 14 percent of the firms offer formal train- ing, up from 5 percent in 2005. Since 2005, the proportion of firms using e-mail has doubled and the proportion of firms that have a website has risen by 67 percent. Twenty-one percent of retail firms reported sales via indirect exports. 20 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 14 Governance and Competitiveness FIN High USA SWE DNK TWN SGP Growth Competitiveness Index NOR CHE r = 0.90 ISL AUS NLD JPN DEU GBR NZL KOR CAN QAT ARE MYS PRT CHL IRL AUT LUX ISR FRA ESP HKG KWT CYP SVN BEL THA MLT TUN CZE BHR SVKZAF HUN LVA JOR LTU GRC ITA BWA CHN IND MEX POL EGY MUS BGR URY JAM SLV COL KAZ GHA TTO ROM BRA HRV NAM CRI AZE IDN RUS TZA ARG PER TUR PHL DZA PAN MAR PAK UKR VNM YUG GEO MDA GMBMKD NGA VEN MOZ UGA MLI MNG GTM HND KEN BOL ECUALB BIH NIC LKA TJK MWI DOM ZWE MDG BGD ETH TMP KHM CMR GUY PRY BEN KGZ TCD Low -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Low Control of Corruption High Source: GCI is based on 2006 data of the World Economic Forum: control of corruption is for 2004, from Kaufmann et al. 2005. supported by other analyses: Afghanistan ranks 162nd of 181 coun- tries in the World Bank's Doing Business 2009 report and 176th of 180 countries in Transparency International's Corruption Perception Index 2008 (annex 4). 2.21 Perceptions regarding governance appear to have deteriorated significantly since 2005. Figure 15 shows that severity perceptions on all governance indicators except corruption declined dramatically between 2005 and 2008. FIGURE 15 Change in Perceptions on Governance Indicators Between 2005 and 2008 Percentage Change in Perceptions from 2005 to 2008 Governanace 350 289 275 300 250 Percentage 200 150 110 91 100 50 ­17 0 ­50 Policy Crime,Theft, Corruption Courts Business Enforcement and Disorder Licensing and Permits 21 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 2.22 Of the governance indicators, policy enforcement14 ranks high- est in the percentage of enterprises that perceive it as a major or very severe constraint. This is true across all sectors (manufacturing, con- struction, retail); for all firm types (registered and unregistered); for all firm size categories except small firms; and for most cities covered by the survey. Overall, 70 percent of the sample considers policy enforcement to be a major or very severe obstacle to their operations. It is noteworthy that its close kin-- macroeconomic instability--did not rate as an issue in 2005.15 2.23 Only 14 percent of respondent firms agreed that laws are inter- preted consistently and predictably by government officials. Across all sectors and cities--and irrespective of firm size, type, or market--at least 70 percent of firms reported a lack of consistency and predictabil- ity (relating to their establishments) on the part of government offi- cials. Overall confidence in the judicial system is also low, with 84 per- cent of Afghan businesses finding the court system corrupt and unfair. 2.24 There has been a marked deterioration since 2005 in the way crime, theft, and disorder are perceived by Afghan firms. In 2005, crime was cited by just 16 percent of the sample as a major problem and ranked outside the top 10 constraints. In 2008, crime ranks third, after policy enforcement and electricity, and is perceived as a serious obstacle by 60 percent of Afghan enterprises. This is corroborated by 14 What is "policy enforcement"? The 2008 Enterprise Survey asked firms whether "political instability" (not "policy enforcement") was a constraint. However, discussions with the government and other stakeholders revealed that this term--especially in view of the fact that it emerged as the number one constraint--could have several sensitive political connotations that were not necessarily implied by the questionnaire. To clarify the issue, 10 firms that reported "political instability" as the biggest obstacle to their business were recontacted to ask what they understood the phrase to mean, and three independent Dari-English bilingual speakers were asked to translate and interpret the phrase. Broadly, the phrase was understood to imply an unstable government or political turmoil leading to frequent policy reversals and weak enforcement of policies. This understanding was dis- cussed in detail with key stakeholders, and a consensus emerged to rename this constraint "policy enforcement," which would imply "state ineffectiveness in designing and implementing consistent policies." It has also been agreed that future enterprise surveys will attempt to unpackage this indica- tor to better understand its exact impact on private sector enterprises. 15 An exact comparison of policy enforcement cannot be made with the 2005 survey, in which respondents were asked about "macroeconomic instability," which may have been interpreted differ- ently. In 2005, only 18 percent of firms ranked macroeconomic instability as a major constraint. In fact, this constraint did not even rank among the top 10 in the 2005 survey. 22 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 panel firms: 22 percent complained about severity in crime in 2005, and 59 percent complained in 2008. Crime ranks among the top four constraints for all sectors, cities, firm sizes types, and for both exporters and non-exporters. 2.25 One in ten surveyed firms reported losses due to theft, robbery, vandalism, or arson (figure 16). Moreover, 36 percent of Afghan enterprises pay for their own security, including equipment, personnel, and professional security services. The average annual cost of security was 140,525 Afghanis (US$2,679). However, security costs for all firms fell from 15 percent of sales in 2005 to just 3 percent in 2008. Six per- cent of surveyed companies also reported making protection payments to government officials or military commanders (figure 17). FIGURE 16 Many Afghan Firms Are Victims of Theft and Vandalism Percentage of Firms that Reported Losses Due to Theft, Robbery, Arson or Vandalism 30 25 24 20 Percentage 20 17 15 13 13 13 10 11 11 11 10 10 8 7 8 9 8 5 1 0 Retail Kandahar Less than 5 Registered Manufacturing Construction Herat Exporter Kabul Mazar-e- Jalalabad 5 to 19 20 to 99 100 or more Unregistered Non-Exporter All Firms Sharif Overall By Sector By Province By Size By Type By Market FIGURE 17 Some Afghan Firms Make Protection Payments to Government Officials or Military Commanders Percentage of Firms that Make Protection Payments 20 20 18 16 14 Percentage 12 11 11 10 8 8 8 8 8 6 7 7 6 6 5 4 3 4 3 2 2 2 0 Manufacturing Construction Non-Exporter Unregistered 100 or more Less than 5 Registered Kandahar Jalalabad Exporter All Firms 20 to 99 5 to 19 Retail Herat Kabul Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 23 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 18 Effect of Crime on Different-Sized Firms Percentage of Firms that Reported Losses due Percentage of Firms that Pay for Security (By Size) to Theft, Robbery, Arson, or Vandalism 25 70 66 20 60 20 50 44 Percentage Percentage 15 13 40 32 10 8 9 30 26 20 5 10 0 0 Less than 5 5 to 19 20 to 99 100 or more Less than 5 5 to 19 20 to 99 100 or more Percentage of Firms that Make Protection Annual (Median) Cost of Security (By Size) Payments (By Size) 160000 150000 25 20 140000 120000 20 120000 Percentage 100000 Afghanis 15 80000 10 8 60000 5 40000 5 2 20000 3360 6000 0 0 Less than 5 5 to 19 20 to 99 100 or more Less than 5 5 to 19 20 to 99 100 or more 2.26 Crime significantly affects larger firms. Medium-sized and large firms tend to perceive crime, theft, and disorder as severe more often than micro and small firms. This perception is validated by the fact that the percentage of larger firms reporting losses from theft is more than double that of micro and small firms. Larger firms are more likely than smaller firms to pay for security or make protection payments, and they incur a higher annual average cost for security (figure 18). 2.27 Nearly 90 percent of firms in Herat cited crime as a major obsta- cle. Many firms in Kabul and Jalalabad also cited crime as a major obstacle. In these three cities, compared with other cities, a larger per- centage of firms had to pay for security. Herat also had the largest per- centage of firms that reported losses as a result of theft, robbery, arson, or vandalism. Mazar-e-Sharif fared best among the five cities, with just 1 percent of firms facing such losses. 2.28 Crime poses a relatively more serious obstacle for the retail sec- tor and for construction firms. Eleven percent of firms in construction and retail (compared with 8% in manufacturing) suffered losses as a 24 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 result of crime, theft, arson, or vandalism. Construction firms were more likely than manufacturing or retail firms to pay for their own security or making protection payments. The median cost of security was also higher for construction firms: 90,000 Afghanis, compared with 24,500 Afghanis for manufacturing and 6,000 Afghanis for retail firms. Crime, theft, and disorder pose a major challenge to medium- sized and large manufacturing firms, micro construction firms, and medium-sized retail firms in particular. 2.29 Exporter firms are relatively more constrained with respect to crime. Exporters are more likely to spend on security and make pro- tection payments as compared to non-exporters. 17 percent of export- ing firms, as opposed to 10 percent of non-exporting firms, also reported losses on account of theft, robbery, arson or vandalism. About 5 percent of the consignment value of exported products is lost in transit owing to theft. 2.30 While perception of corruption has improved since 2005, the 2008 quantitative indicators for corruption indicate that it is still a serious issue. In their dealings with public officials, Afghan enterprises pay over 2 percent of their sales as bribes to "get things done." Afghan businesses were asked whether an informal gift was expected or requested in the course of their business transactions and the responses are indicative of the extent of corruption (figure 19). Firms are more likely to report an informal gift for an electricity connection, construction permit, or water connection. Corruption seems to be alive and well in Afghanistan, as also reported in the Transparency International 2009 rankings (annex 4). FIGURE 19 Informal Gifts Are Common Practice in Afghanistan Percentage of Firms that Reported Informal Gift For: 35 32 31 29 30 30 26 25 23 20 15 9 10 7 5 0 Registration An Operating An Import An Electricity A Water A Telephone A Construction During Tax Process License License Connection Connection Connection Related Permit Inspections 25 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 2.31 Corruption exists throughout the country, but its impact is uneven across firms of different sizes and across different cities. Larger firms generally perceive corruption as a bigger issue, although many small firms report informal gifts or payments in their dealings with public officials, especially for the registration process, securing operating licenses, or obtaining a water connection. Over half the firms in Herat, Jalalabad, and Kabul reported corruption as a severe constraint. Over 20 percent of firms in Herat, Kandahar, and Mazar- e-Sharif (compared with just 4% of firms in Kabul and 6% of those in Jalalabad) reported that an informal gift was expected or requested for the registration process. For an operating license, two out of three firms in Mazar-e-Sharif and two out of five in Jalalabad report paying bribes. To secure an import license, about half of the firms in Mazar- e-Sharif and Herat, and a third of those in Jalalabad, pay a bribe or make a gift. 2.32 Almost all quantitative indicators of corruption suggest that this factor is a more severe constraint for unregistered firms than for registered firms.16 A larger percentage of unregistered firms than reg- istered firms pay bribes in most kinds of dealings with public officials (figure 20). FIGURE 20 Corruption Indicators for Registered and Unregistered Firms Percentage of Firms that Reported Informal Gift For: (By Firm Type) 70 64 60 56 50 43 Percentage 41 40 29 30 24 21 21 20 15 12 11 10 6 0 An Operating An Import A Water A Telephone A Construction During Tax License License Connection Connection Related Permit Inspections Registered Unregistered 16 This finding is in contrast with the perception data, in which corruption is perceived as a major or very severe constraint by 50 percent of registered firms and just 33 percent of unregis- tered firms. Corruption is also ranked higher for registered firms (fourth) than for unregistered firms (sixth). 26 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 21 Exporting Firms Face Higher Pressures From Corruption Percentage of Firms that Reported Informal Gift for (By Market): 40 36 35 31 31 31 30 30 23 25 Percentage 25 20 15 13 10 9 5 4 0 Registration An Operating An Import An Electricity During Tax Process License License Connection Inspections Exporter firms Non-exporter firms 2.33 Corruption is a very serious constraint for construction firms. Sixty percent of enterprises in the construction sector perceive corrup- tion as posing a severe obstacle to their operations--much higher than in manufacturing (43%) or retail (48%). An analysis by size of con- struction firms reveals that small firms are relatively better placed. Less than half of the small construction firms complained about corrup- tion; in all other size categories, the figure was over 70 percent. Even though perceptions relating to corruption are better for manufactur- ing, 60 percent of medium-sized manufacturing firms cite corruption as a serious obstacle. In retail, there appears to be a direct correlation between the size of the firm and the perception of constraint severity with respect to corruption. 2.34 Exporters are more likely to make an informal gift or payment to get things done, except for registration and procuring an import license (figure 21). Exporters also pay a smaller proportion of the con- tract value as informal payment in order to secure contracts. GETTING FACTOR MARKETS TO WORK BETTER 2.35 Factor markets have seen mixed improvements since 2005 (figure 22). The need to get factor markets to work better is another key message from this survey. We look at the traditional factor mar- kets of capital, labor, and land. We define land as serviced industrial land, with electricity and water as the fundamental services. 27 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 22 Change in Perceptions of Factor Market Indicators Between 2005 and 2008 Percentage Change in Perceptions from 2005 to 2008 Factor Markets 140 125 120 100 Percentage 80 60 40 20 11 ­25 ­31 3 ­5 0 Telecommunications Electricity Educated Workforce Inadequately ­20 Transport Access to Finance ­40 Access to Land SERVICED LAND 2.36 In 2008, 41 percent of surveyed firms reported access to land as a major or very severe constraint to their business operations. While perceptions relating to the severity of this constraint have improved since 2005--when access to land ranked second after electricity--it is now perceived as the fifth most important constraint. The only major deviations from the overall ranking are seen in smaller cities, where as many as 65 percent of firms complain about land; and, conversely, in Jalalabad, where only 23 percent complain. Medium-sized and large firms (compared with micro and small firms) and the construction sector (compared with other sectors) are relatively more constrained by land access. 2.37 Additional land is not easy to obtain--68 percent of the respon- dents who tried to acquire new land in the past three years failed to do so. In fact, the figure for Kabul is as high as 83 percent. Overall, it took an average of 192 days to obtain land. Inability to acquire land can be attributed to several factors, primarily because the government did not want to sell it. One in five firms also found buying new land too expen- sive. Failure to obtain zoning approval, failure to get a secure title, and mafia-related issues were also cited as reasons for the inability to acquire land (figure 23). The fact that 4 percent of respondents 28 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 23 Reasons for Inability to Acquire Additional Land Reasons why Firms Failed to Acquire Additional Land (% Firms) Govt. did not want to sell 19% Could not get zoning 28% approval Could not get secure title 4% Govt. could not provide infrastructure Too expensive 22% 11% Official, mafia, commander intervention 8% Other 8% reported being unable to acquire land because of mafia or military commander intervention may indicate a correlation between access to land and crime/security; however, the survey does not have enough information to confirm or refute this hypothesis. 2.38 It took firms an average of 88 days to obtain a construction permit. Among Afghan businesses that applied for construction- related permits, 31 percent reported that an informal gift was expected or requested. 2.39 Obtaining additional land appears to be less of a challenge in Kandahar, although it takes longer. Of the firms that attempted to acquire land in this region, 81 percent succeeded. As a matter of fact, Mazar-e-Sharif, Kandahar, and Jalalabad reported a lower percentage of firms that cited land access as a major or very severe constraint com- pared with the overall average of 41 percent. However, it can take an aver- age of 247 days to obtain land in Kandahar, compared with the 4 days reported by one firm in Mazar-e-Sharif.17 Kabul reported the highest percentage of firms that failed to acquire additional land; the main rea- son was the unwillingness of the government to sell. Kabul also fared 17 In Mazar-e-Sharif, only 4 out of 142 firms applied for additional land. Of these, only 1 firm suc- ceeded in acquiring it. The other three firms found the land to be too expensive. Only one firm applied for a construction permit. 29 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 24 City Variations on Access to Land Indicators Access to Land (a) Access to Land (b) 90 81 400 80 350 70 60 300 60 247 Percentage 50 250 205 Days 40 38 36 36 200 25 29 150 30 17 100 87 20 64 59 53 10 50 4 19 16 0 0 0 0 % Firms that succeeded in % Firms that reported informal Av. No. of days to obtain Av. No. of days to obtain acquiring additional land gift for construction permit additional land construction permit Kabul Herat Kandahar Kabul Herat Kandahar Mazar-e-Sharif Jalalabad Mazar-e-Sharif Jalalabad worst in the number of days needed to obtain a construction permit. Jalalabad was worst in terms of corruption related to construction per- mits (figure 24).18 2.40 The construction sector is relatively more constrained than others with respect to land access. The severity of this constraint is reported to be higher for medium-sized and large firms than for small and micro firms. A smaller proportion of construction firms than manufacturing or retail firms succeeded in acquiring new land. Failure to acquire land because of an inability to get zoning approvals was also cited by more construction firms than manufacturing or retail firms 2.41 Electricity is a constraint for most firms, despite some improve- ments. This result is virtually unchanged from the last survey. In 2008, 66 percent of the sample cited electricity as a major or very severe obstacle to doing business. In 2005, the figure was 64 percent. Although several quantitative indicators for electricity show an improvement from 2005, it remains a serious obstacle constraining business operations. 2.42 Firms are either unconnected or suffer from unreliable connec- tions. About 20 percent of surveyed firms do not have access to the grid. The problem is more acute for retail firms than for other sectors; 18 There were only a small number of responses to these questions, therefore further study is rec- ommended to properly dimension this issue. 30 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 25 Many Firms Have No Access to a Public Supply of Electricity Percentage of Firms with no Electricity 33 35 30 31 30 27 27 Percentage 23 22 22 25 21 19 20 20 19 14 14 15 12 7 8 10 5 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Non-Exporter Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market for Kabul and Jalalabad than for other cities;19 for micro firms than larger firms; for unregistered firms than registered firms; and for non- exporters than exporters (figure 25). 2.43 Power outages are a regular feature for Afghan businesses. Among firms having access to the grid, 73 percent reported suffering power outages, with an average of 23 outages per month, each lasting about 11 hours. However, the situation has improved substantially since 2005. In the 2005 survey, firms reported having power for an average of just six and a half hours a day. For the panel firms, outage duration dropped from 20 hours in 2005 to 8 hours in 2008. Despite this, firms in 2008 reported output losses from power outages averag- ing 9 percent of sales. There is enormous dependence on privately gen- erated electricity. Seventy-five percent of Afghan enterprises report having a generator, with 77 percent of electricity coming from private sources of power figure 26). This is costly for firms--the industrial tar- iff for getting electricity from the grid is about US$0.20 per KW hour, while private generation can cost upwards of US$0.30 per KW hour.20 19 When our survey ended in December 2008, Kabul received a total of 30MW of power per hour. In January 2009, a power line constructed between Kabul and Uzbekistan became active, delivering an extra 40MW of power per hour to Kabul. The availability of power is expected to increase further. Under a project funded by the Asian Development Bank (ADB), a transmission line will be constructed from Kabul to Jalalabad over the next two years to provide power to Jalalabad. 20 These are approximate estimates; the cost varies by region. 31 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 26 Most Enterprises Rely on Self-Generated Power Percentage of Electricity from Generator 100 100 92 88 90 83 84 80 77 77 76 79 75 77 80 72 73 67 70 59 Percentage 60 50 42 40 30 20 10 0 All Firms Manufacturing Non-Exporter Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Mazar-e- Overall By Sector By Province Sharif By Size By Type By Market *Note: The 100 percent figure for Retail is based on responses obtained from only 2 retail firms in the sample. 2.44 Perceptions relating to electricity are worse for manufacturing firms than for construction and retail firms. Managers in manufac- turing firms are more likely to perceive electricity as a serious con- straint compared with managers in construction and retail activities. In all three sectors, micro firms are less constrained by electricity problems. 2.45 Water is not as serious an issue as electricity for Afghan firms. About 8 percent of the firms reported experiencing insufficient water supply. Public sources of water account for 17 percent of the water used in the production process. The main problem with regard to water as a factor input is corruption--30 percent of surveyed firms said that an informal gift or payment was expected or requested for a water connection. This problem appears to be more acute for the con- struction sector, in Jalalabad and Kabul, for smaller firms, and for unregistered firms. 2.46 Lack of serviced land is a serious problem. Firms identified elec- tricity as the second biggest problem and availability of land as the fifth. While steps are being undertaken to provide electricity for the entire country, tackling the issue of serviced land would deal with both these issues in a reasonably quick and practical way. 32 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 ACCESS TO FINANCE 2.47 Access to finance was perceived as a major or very severe con- straint to doing business by 38 percent of the sample in 2008, com- pared with 51 percent in 2005. The corresponding figures for panel firms are 40 percent for 2008 and 50 percent for 2005. Despite improvement since 2005, Afghanistan performs poorly relative to comparator countries on quantitative indicators of access to finance (annex 4). 2.48 Afghanistan has made some progress in terms of financial deepening, but there is a long way to go. Only 51 percent of enter- prises reported having a bank account, although this is a marked improvement since 2005, when only 30 percent had a bank account. Exporters (compared with non-exporters), registered firms (com- pared with unregistered firms), larger companies (compared with smaller companies), and firms in the construction industry (com- pared with manufacturing and retail) are more likely to have a bank account (figure 27). Moreover, 97 percent of enterprises that do have bank accounts are very satisfied or quite satisfied with the services provided by their banks--this sentiment is expressed across sectors, cities, firm sizes, firm types, and markets. Thirty-four percent of enterprises that use banking services also reported having an over- draft facility. However, electronic banking mechanisms need strengthening--only 1 percent, 5 percent, and 1 percent of the respondents reported using ATM cards, credit cards, and payment services, respectively. FIGURE 27 Percentage of Afghan Firms That Have a Bank Account Percentage of Firms that have a Bank Account 100 86 90 80 80 67 69 66 Percentage 70 51 56 56 58 60 45 49 50 45 51 50 31 40 26 28 30 20 10 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Non-Exporter Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 33 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 28 Percentage of Firms Accessing Formal Credit Percentage of Firms with Loan from a Financial Institution 14 13 12 11 9 Percentage 10 8 6 6 6 6 5 5 4 4 5 4 4 4 2 2 1 2 2 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Non-Exporter Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 2.49 Only 5 percent of Afghan enterprises have a loan with a financial institution (figure 28). This finding is reinforced by the World Bank's Doing Business 2009 report, which ranks Afghanistan 178th of 181 countries in "getting credit." A major proportion of these loans is provided by private commercial banks (45%), followed by non-banking financial institutions (NBFIs) (29%). Nine in 10 surveyed firms had to supply collateral against their loans; just over half of the firms reported using personal assets, while just under half used land as collateral. 2.50 Close to half of all firms felt the need to access formal credit but did not do so. The reasons varied widely, as illustrated in figure 29. The finding that 51 percent of firms reported having no need for formal credit warrants further research. FIGURE 29 Reasons for Not Accessing Formal Credit Main Reason for not Applying for a Loan (% Firms) 60 51 50 Percentage 40 30 26 20 8 10 3 4 3 4 1 0 No Need Application Int. Rate not High Loan Size May not be Islam Other Procedure Favorable Collateral Insufficient Approved Complex 34 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 2.51 Working capital is overwhelmingly obtained from non-formal sources of finance. Firms across all sectors, cities, size categories, and firm types report financing between 80 and 90 percent of their work- ing capital from retained earnings. Dependence on banks and NBFIs is just 2 percent. 2.52 Only 7 percent of fixed-asset purchases are made out of owners' contributions or issuance of new equity shares. This is an improve- ment over the 3 percent figure reported in 2005. Equity financing is relatively more important in manufacturing and retail than in con- struction; for smaller firms; and for registered firms compared with unregistered firms. Retained earnings are the primary source of financing, accounting for 85 percent of all fixed-asset purchases. Over- all, dependence on banks, NBFIs, credit purchases, and other sources (friends, relatives, moneylenders) is miniscule. In fact, no micro firm reported purchasing fixed assets through financing from banks, NBFIs, or on credit. 2.53 Between 2005 and 2008, financing of new investments from retained earnings for all firms increased from 78 percent to 85 per- cent. While equity financing has shown a slight improvement, little progress is seen with regard to bank credit (figure 30). 2.54 Perceptions relating to access to finance are better for con- struction firms than for manufacturing or retail firms. Finance indicators reveal a mixed picture on access to and availability of financing across sectors. Four in five construction firms have a bank account--a higher proportion than that of firms in other sectors. FIGURE 30 Source of Financing for New Investments Financing of New Investments from: 100 85 78 80 Percentage 2005 2008 60 40 20 2.5 7 1 2 0 Retained Earnings Equity Banks 35 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 However, construction firms are least likely to have a loan with a financial institution (figure 28). Construction reports a higher per- centage of sales made on credit, followed by manufacturing and then retail. 2.55 The quantitative indicators for access to finance place unregis- tered firms in a weaker position. Although the difference in percep- tions is marginal with respect to access to financing, only 28 percent of unregistered firms reported having a bank account, compared with 66 percent of registered firms. Among the unregistered firms that do have a bank account, ATM or credit card use was nil. Rela- tively fewer unregistered businesses have an overdraft facility than registered firms. Furthermore, a higher proportion of inputs are pur- chased on credit by unregistered firms than by registered firms. The proportion of sales made on credit, however, is higher for registered firms (figure 31). 2.56 Unregistered firms are more likely to depend on NBFIs for a loan. Half of all unregistered firms reported obtaining a loan from an NBFI, compared with 12 percent of registered enterprises, which relied more on private commercial banks or state banks for loans. Virtually every respondent in the unregistered sector was required to provide some collateral, compared with 80 percent of registered firms. The majority of unregistered firms used personal assets (e.g., a house) as collateral, while most registered firms used land. FIGURE 31 Transactions on Credit by Firm Type Transactions on Credit 40 34 35 30 30 28 25 20 19 15 10 5 0 Percentage of Inputs Percentage of Sales Purchased on Credit Made on Credit Registered Unregistered 36 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 2.57 A comparison of finance indicators with other countries reveals a mixed picture. Afghanistan has a larger percentage of firms with a bank account than Yemen does, but it has a long way to go to achieve, for example, the 94 percent figure of Jordan (figure 32). Afghanistan also has the lowest percentage of firms with loans from a financial institution (figure 33). With regard to new investments from banks and other financial institutions, Afghanistan, Jordan, and the West Bank and Gaza are comparable (figure 34). And even though Yemeni FIGURE 32 Country Statistics on Firms With a Bank Account Percentage of Firms that have a Bank Account 100 94 80 63 Percentage 60 51 40 20 5 0 Afghanistan Jordan Pakistan Yemen FIGURE 33 Country Statistics on Loans From Financial Institutions Percentage of Firms with Loan from a Financial Institution 50 46 45 40 35 34 Percentage 30 30 25 20 18 17 15 14 13 10 9 5 5 0 Afghanistan Bhiopia Tajikistan Jordan West Bank Yemen Egypt Syria Pakistan and Gaza FIGURE 34 Country Statistics on Share of New Investments From Formal Sources Share of New Investments from Banks and Other Financial Institutions 12 11 10 9 Percentage 8 6 4 4 3 3 2 2 0 Afghanistan Egypt Yemen Syria Jordan West Bank & Gaza 37 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 firms are less likely to have a bank account, they are more likely to have loans or new investment from banks and financial institutions. 2.58 While there has been improvement, access to formal finance remains a problem. Access to finance was the single biggest obstacle reported by Afghan businesses, and it is generally recognized as a major problem area. The reform agenda is complicated and will take years to implement. In the meantime, it would be useful for policy- makers to develop practical solutions to improve access to finance for the underserved. LABOR MARKETS 2.59 Rigidities in labor markets or in the availability of skills are not generally perceived to be a problem. While an inadequately educated workforce does not figure among the top investment climate con- straints for Afghanistan, a comparison across sectors shows that con- struction firms are relatively more constrained than manufacturing and retail with respect to skills and education of the workforce. 2.60 While the majority of workers across the board is literate, less than one in three has a secondary education. Only 1 of the 220 manu- facturing firms that reported the level of educational attainment of its workers said "college education." While more than 80 percent of firms do not perceive lack of skills as an issue, a number of them--especially large and registered firms--report being constrained in terms of the availability of skilled professionals and experts (figure 35). FIGURE 35 Some Afghan Firms Face Skill Shortages Percentage of Firms that Need Skills that cannot be Found or are Difficult to find in Afghanistan 45 41 40 35 Percentage 26 27 28 30 21 24 21 23 25 18 17 18 20 12 14 11 15 9 9 10 5 5 0 Retail Kandahar Herat Jalalabad Less than 5 Registered All Firms Manufacturing Construction 5 to 19 20 to 99 100 or more Unregistered Exporter Non-Exporter Kabul Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 38 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 INNOVATIVE AND PRODUCTIVITY- ENHANCING BEHAVIORS 2.61 A key message of this report is that Afghan firms are not suffi- ciently exhibiting innovative and productivity-enhancing behaviors. While the survey indicates that the high growth witnessed by Afghanistan's private sector may continue in the short term, a lack of innovative and productivity-enhancing behaviors on the part of Afghan firms could adversely impact the growth pattern of the private sector in the medium and long term. Some of these behaviors are examined below in terms of (a) education and training indicators of the workforce, especially IT skills; (b) efforts by businesses to improve their technology or product quality; and (c) expansion plans of Afghan businesses. 2.62 About 14 percent of the surveyed firms offer formal training (figure 36). This is an improvement over the 5 percent figure reported in 2005. There is a steady increase with firm size in the per- centage of firms that train their permanent full-time employees. For- mal training in computers is provided by over half of all surveyed firms. Registered firms do better on training and report a marginally higher staff literacy rate. 2.63 Computer and Internet use is weak. This is particularly true for micro and small firms, unregistered firms, and manufacturing and FIGURE 36 Firms That Provide Formal Training for Employees Percentage of Firms with Formal Training Programs for their Permanent, Full-Time Employees 40 35 33 35 30 Percentage 25 22 20 18 16 14 14 12 13 13 15 10 9 10 9 10 8 5 1 3 0 All Firms Retail Less than 5 Manufacturing Construction Kandahar 100 or more Registered Herat Jalalabad 5 to 19 20 to 99 Unregistered Exporter Non-Exporter Kabul Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market 39 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 37 Computer and Internet Use Varies Widely Among Afghan Firms Percentage of Enterprises that have: 80 70 60 Percentage 50 40 30 20 10 0 Retail Less than 5 Manufacturing Construction Herat Kandahar Mazar-e- Jalalabad 5 to 19 20 to 99 100 or more Registered Unregistered Exporters Non- Exporters All Firms Kabul Sharif Overall By Sector By Province By Size By Type By Market Use email Have own website Have internet connection FIGURE 38 Percentage of Firms That Use Foreign-Licensed Technology Percentage of Firms that have Technology Licensed from a Foreign-Owned Company 30 26 25 Percentage 20 15 11 8 8 9 8 9 10 6 6 7 6 5 2 0 0 0 0 Retail Less than 5 Manufacturing Construction Herat Kandahar Mazar-e- Jalalabad 5 to 19 20 to 99 100 or more Registered Unregistered Exporter Sharif Non-Exporter All Firms Kabul Overall By Sector By Province By Size By Type By Market retail sectors (figure 37). There has been some improvement, however, since 2005: the proportion of firms that use e-mail has doubled and the proportion that have their own website has risen by 67 percent. Internet connectivity is a challenge: over half of all responding firms reported interruptions in Internet service, lasting an average of 50 hours (median) with a frequency of 13 interruptions each month. Very few Afghan firms report using technologies licensed from foreign-owned companies (figure 38). 2.64 Afghan firms primarily produce for the domestic market (figure 39). A larger proportion of retail firms and a higher incidence of Jalal- abad-based firms applied for an import license. Exposure to interna- tional markets promotes productivity enhancing behaviors such as 40 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 FIGURE 39 Percentage of Afghan Firms That Applied for an Import License Percentage of Firms that Applied for an Import License 45 43 40 35 Percentage 30 27 25 19 21 20 16 18 18 13 15 13 13 15 9 10 8 6 5 4 3 5 0 All Firms Manufacturing Construction Retail Kabul Herat Kandahar Jalalabad Less than 5 5 to 19 20 to 99 100 or more Registered Unregistered Exporters Non- Exporters Mazar-e- Sharif Overall By Sector By Province By Size By Type By Market FIGURE 40 Percentage of Afghan Firms With Quality Certification Percentage of Firms with an Internationally Recognized Quality Certification 12 10 10 10 10 8 Percentage 8 7 6 5 5 5 5 4 4 4 4 3 2 2 2 2 1 0 All Firms Kabul Jalalabad 100 or more Non-Exporter Overall by Province acquiring quality certifications. However, the incidence of this is very low in Afghanistan (figure 40). CONCLUSIONS 2.65 The survey findings in general validate the Afghan government's current reform program. The government is making efforts to (1) improve governance through better business regulation and stronger institutions, and (2) strengthen factor markets by improving access to finance, providing industrial land, and addressing infrastructure and 41 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 human capital issues. More efforts are required, as the survey reports little measurable progress. 2.66 Within the cluster of reform efforts, the one striking finding of the survey is in the area of policy enforcement. The matter has been discussed in detail with stakeholders, and many anecdotes emerged of policy reversals and failure to uphold commitments with respect to land access or licenses. As noted earlier, investors will not tolerate a lack of predictability; it is a key reason why Afghanistan has not attracted investment despite the potential for attractive returns and the existence of some relatively secure locations to do business. 2.67 The survey and the associated analysis also identify a group of emerging issues that require some attention. These include the interlinked issues of competitiveness, innovation, and diversifica- tion. Despite strong business growth, policymakers should be con- cerned about the recent lack of new firms, especially foreign firms, entering the market. The scarcity of new entrants and the lack of innovative behavior by firms are strong indicators of a lack of com- petitiveness; these areas warrant additional research. Another con- cern is the undiversified industrial sector, which is overwhelmingly (approximately 95%) linked to the agriculture and agro-processing sectors. 2.68 The strong growth also needs deeper evaluation. A thorough understanding of the key drivers of growth will be helpful for policy- makers. In our experience, no country that has weak governance, poor factor markets, and non-innovative firms achieves long-term sustain- able growth. The transition from what seems to be a catch-up and aid- driven demand growth to a more normal growth pattern requires care- ful planning. 2.69 The following chart lists the issues discussed in this report, sum- marizes the reform status, and suggests some next steps. The next steps include further analytic work, dialogue, and strengthening the ongoing reform efforts of the government. We suggest that the enterprise sur- vey be repeated regularly to assess the impact of government policies on private sector growth and productivity. 42 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 Afghanistan Enterprise Survey--Review of Findings and Next Steps Issue Current Status Next Steps Overall Findings The quality and sustainabil- Very few growth diagnostics Carry out a "sources of growth" ity of current levels of are available study. growth is not clear. Carry out an economic diversification study to explore potential areas and identify any policy, regulatory, or structural impediments to their development. The investment climate in No regional or city-level Undertake a dialogue to Afghanistan shows important economic development review local economic regional variations. strategies are available. development opportunities. Competitiveness challenges The topic of competitiveness Carry out a competitiveness exist. has not received much study that looks at firm entry attention. and exit. Some topics were not These topics include rural Ensure that these topics are adequately covered in this enterprises located outside the properly covered in future survey. 10 surveyed cities, gender, surveys. worker skills, and the pre- dictability problem. Governance Government policy and There is a general awareness of Undertake a high-level regulatory predictability are the need for predictability, but dialogue with stakeholders very weak. the importance of the issue is on this topic. not well understood by policymakers and there is no systematic approach to address it. Corruption is widespread This topic is being addressed, Continue to monitor and and shows no signs of but measures in place appear report on the impact of abating. inadequate to reduce corruption on the private corruption. sector. Undertake a dialogue to determine whether the private sector itself could take measures to manage corruption. Crime, theft, arson, and This issue is receiving Continue to monitor and vandalism problems are substantial attention and report on the impact of increasingly perceived as significant international these problems on the constraints. support being made available private sector. to help. Business regulation is weak. A substantial amount of work Continue to monitor and is being done on this topic. report on the impact of weak regulation on the private sector. (continued) 43 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 Issue Current Status Next Steps Court effectiveness is poor Work is being done to Develop a deeper under- but not rated as a constraint, encourage judicial reform. No standing of whether and which suggests the use of work is being done on parallel how informal measures are parallel informal informal mechanisms that deal used to resolve disputes and mechanisms. with commercial and financial establish property rights. disputes. Factor Market Accessing Land is difficult In parallel with efforts to Continue to monitor and reform the overall land regis- report on the impact on the Accessing serviced land is tration system, there are plans private sector. very difficult to develop various types of special economic zones (industrial parks etc) throughout the country. Electric power in the There is a substantial electricity Continue to monitor and quantities and quality infrastructure development report on the impact on the needed is not available program. private sector. In general infrastructure There are development Continue to monitor and needs to be strengthened. programs in all infrastructure report on the impact of this sectors. issue on the private sector. Access to finance is A number of important efforts Carry out financial sector constrained. in the financial sector are update. beginning to produce results, Continue to monitor and including a number of programs report on the impact of this designed to increase the issue on the private sector. availability of finance. Human resource skills A number of vocational Undertake a dialogue to are weak programs are under way, develop a private sector­led but their impact has yet to approach to vocational be felt. training. Continue to monitor and report on the impact of this issue on the private sector. Innovation There are a number of This topic has not received any Undertake a dialogue to indicators of weak firm- attention, with the exception develop an integrated level innovation. of export development. industrial innovation policy. Start monitoring and reporting on the impact of this issue on the private sector. 44 ANNEX 1. AFGHANISTAN AT A GLANCE Key Development Indicators South Low Afghanistan Asia income Age distribution, 2007 (2007) Male Female Population, mid-year (millions) 27.1 1,520 1,296 75­79 Surface area (thousand sq. km) 652 5,140 21,846 Population growth (%) 3.9 1.4 2.1 60­64 Urban population (% of total population) 24 29 32 45­49 GNI (Atlas method, US$ billions) 10.1 1,339 749 30­34 GNI per capita (Atlas method, US$) 370 880 578 15­19 GNI per capita (PPP, international $) 920 2,537 1,500 0­4 GDP growth (%) 13.5 8.5 6.5 30 20 10 0 10 20 30 GDP per capita growth (%) 9.2 6.9 4.3 percent (most recent estimate, 2000­2007) Poverty headcount ratio at $1.25 a day (PPP, %) .. 40 .. Under-5 mortality rate (per 1,000) Poverty headcount ratio at $2.00 a day (PPP, %) .. 74 .. Life expectancy at birth (years) 43 64 57 Infant mortality (per 1,000 live births) 165 62 85 300 Child malnutrition (% of children under 5) 33 41 29 250 200 Adult literacy, male (% of ages 15 and older) 43 70 72 Adult literacy, female (% of ages 15 and older) 13 46 50 150 Gross primary enrollment, male (% of age group) 126 1 111 100 100 Gross primary enrollment, female (% of age group) 75 104 89 50 Access to an improved water source (% of population) 22 87 68 0 1990 1995 2000 2006 Access to improved sanitation facilities (% of population) 30 33 39 Afghanistan South Asia a Net Aid Flows 1980 1990 2000 2007 (US$ millions) Net ODA and official aid 33 122 136 3,000 Growth of GDP and GDP per capita (%) Top 3 donors (in 2006): United States 2 56 2 1,404 20 United Kingdom 1 2 13 246 European Commission 0 2 18 221 15 Aid (% of GNI) 0.9 .. 16.4 31.9 10 Aid per capita (US$) 2 10 7 115 5 Long-Term Economic Trends 0 95 05 Consumer prices (annual % change) .. .. ­11.8 12.7 GDP implicit deflator (annual % change) .. .. .. 9.4 GDP GDP per capita Exchange rate (annual average, local per US$) 44.1 50.6 67.7 37.6 Terms of trade index (2000 = 100) .. .. .. .. 1980­90 1990­2000 2000­07 (average annual growth %) Population, mid-year (millions) 13.9 12.7 20.7 27.1 ­1.0 4.9 3.8 GDP (US$ millions) 3,642 .. 2,462 11,627 .. .. 11.5 (% of GDP) Agriculture .. .. 49.8 36.1 .. .. .. Industry .. .. 20.1 24.5 .. .. .. Manufacturing .. .. 15.1 14.9 .. .. .. Services .. .. 30.1 39.4 .. .. .. Household final consumption expenditure .. .. 97.5 110.8 .. .. .. General gov't final consumption expenditure .. .. 7.8 9.7 .. .. .. Gross capital formation 13.9 .. 28.1 28.2 .. .. .. Exports of goods and services 10.8 .. 32.9 14.0 .. .. .. Imports of goods and services 13.9 .. 66.2 62.7 .. .. .. Gross savings .. .. 24.7 26.5 Note: 2007 data are preliminary estimates. This table was produced from the Development Economics LDB database. * The diamonds show four key indicators in the country (in bold) compared with its income-group average. If data are missing, the diamond will be incomplete. 45 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 Balance of Payments and Trade 2000 2007 Governance indicators, 2000 and 2007 (US$ millions) Total merchandise exports (fob) 69 755 Voice and accountability Total merchandise imports (cif) 1,697 3,280 Net trade in goods and services ­1,363 ­3,144 Political stability Workers' remittances and Regulatory quality compensation of employees (receipts) .. .. Rule of law Current account balance ­150 ­70 as a % of GDP ­3.7 ­1.1 Control of corruption Reserves, including gold 425 1,662 0 25 50 75 100 2007 Country's percentile rank (0­100) Central Government Finance higher values imply better ratings 2000 (% of GDP) Current revenue (including grants) .. 11.1 Source: Kaufmann-Kraay-Mastruzzi, World Bank Tax revenue .. .. Current expenditure .. 9.9 Technology and Infrastructure 2000 2007 Overall surplus/deficit .. 0.1 Paved roads (% of total) 13.3 23.7 Highest marginal tax rate (%) Fixed line and mobile phone Individual .. .. subscribers (per 1,000 people) 0 18 Corporate .. .. High technology exports (% of manufactured exports) .. .. External Debt and Resource Flows Environment (US$ millions) Total debt outstanding and disbursed .. 1,771 Agricultural land (% of land area) 58 58 Total debt service .. 9 Forest area (% of land area) 1.6 1.3 Debt relief (HIPC, MDRI) 546 .. Nationally protected areas (% of land area) .. 0.3 Total debt (% of GDP) .. 18.9 Freshwater resources per capita (cu. meters) .. 2,194 Total debt service (% of exports) .. .. Freshwater withdrawal (% of internal resources) 42.3 .. Foreign direct investment (net inflows) .. .. CO 2 emissions per capita (mt) 0.04 0.03 Portfolio equity (net inflows) .. 0 GDP per unit of energy use (2005 PPP $ per kg of oil equivalent) .. .. Composition of total external debt, 2006 Energy use per capita (kg of oil equivalent) .. .. Short-term, 11 IBRD, 0 Private, 0 IDA, 358 World Bank Group portfolio 2000 2007 IMF, 0 (US$ millions) IBRD Bilateral, 995 Total debt outstanding and disbursed ­ 0 Other multi- Disbursements ­ 0 lateral, 408 Principal repayments ­ 0 Interest payments ­ 0 US$ millions IDA Total debt outstanding and disbursed ­ 411 Disbursements ­ 56 Private Sector Development 2000 2008 Total debt service ­ 7 Time required to start a business (days) ­ 9 IFC (fiscal year) Cost to start a business (% of GNI per capita) ­ 59.5 Total disbursed and outstanding portfolio 0 9 Time required to register property (days) ­ 250 of which IFC own account 0 9 Disbursements for IFC own account 0 1 Ranked as a major constraint to business 2000 2007 Portfolio sales, prepayments and (% of managers surveyed who agreed) repayments for IFC own account 0 0 n.a. .. .. n.a. .. .. MIGA Gross exposure 0 77 Stock market capitalization (% of GDP) .. .. New guarantees 0 75 Bank capital to asset ratio (%) .. .. Note: This table was produced from the Development Economics LDB database. 46 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 With selected targets to achieve between 1990 and 2015 (estimate closest to date shown, +/­ 2 years) Afghanistan Goal 1: halve the rates for extreme poverty and malnutrition 1990 1995 2000 2007 Poverty headcount ratio at $1.25 a day (PPP, % of population) .. .. .. .. Poverty headcount ratio at national poverty line (% of population) .. .. .. .. Share of income or consumption to the poorest qunitile (%) .. .. .. .. Prevalence of malnutrition (% of children under 5) .. .. .. 32.9 Goal 2: ensure that children are able to complete primary schooling Primary school enrollment (net, %) 25 .. .. .. Primary completion rate (% of relevant age group) .. 25 .. 38 Secondary school enrollment (gross, %) 16 .. 13 19 Youth literacy rate (% of people ages 15­24) .. .. 34 .. Goal 3: eliminate gender disparity in education and empower women Ratio of girls to boys in primary and secondary education (%) 54 .. .. 56 Women employed in the nonagricultural sector (% of nonagricultural employment) 18 .. .. .. Proportion of seats held by women in national parliament (%) 4 .. .. 27 Goal 4: reduce under-5 mortality by two-thirds Under-5 mortality rate (per 1,000) 260 257 257 257 Infant mortality rate (per 1,000 live births) 168 165 165 165 Measles immunization (proportion of one-year olds immunized, %) 20 41 35 68 Goal 5: reduce maternal mortality by three-fourths Maternal mortality ratio (modeled estimate, per 100,000 live births) .. .. .. 1,800 Births attended by skilled health staff (% of total) .. .. 12 14 Contraceptive prevalence (% of women ages 15­49) .. .. 5 10 Goal 6: halt and begin to reverse the spread of HIV/AIDS and other major diseases Prevalence of HIV (% of population ages 15­49) .. .. .. .. Incidence of tuberculosis (per 100,000 people) 248 247 208 161 Tuberculosis cases detected under DOTS (%) .. 3 15 66 Goal 7: halve the proportion of people without sustainable access to basic needs Access to an improved water source (% of population) .. 21 21 22 Access to improved sanitation facilities (% of population) .. 32 30 30 Forest area (% of total land area) 2.0 .. 1.6 1.3 Nationally protected areas (% of total land area) .. .. .. 0.3 CO2 emissions (metric tons per capita) 0.2 0.1 0.0 0.0 GDP per unit of energy use (constant 2005 PPP $ per kg of oil equivalent) .. .. .. .. Goal 8: develop a global partnership for development Telephone mainlines (per 100 people) 0.3 0.2 0.1 0.3 Mobile phone subscribers (per 100 people) 0.0 0.0 0.0 17.2 Internet users (per 100 people) 0.0 .. 0.0 2.1 Personal computers (per 100 people) .. .. .. 0.4 Education indicators (%) Measles immunization (% of 1-year olds) ICT indicators (per 1,000 people) 75 100 20 50 75 15 50 10 25 25 5 0 2000 2002 2004 2006 0 0 1990 1995 2000 2006 2000 2002 2004 2006 Primary net enrollment ratio (..) Afghanistan South Asia Fixed + mobile subscribers Ratio of girls to boys in primary & secondary education Internet users 47 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 ANNEX 2. THE ENTERPRISE SURVEY SAMPLE 97. Sample size. The sample target was 1,000 companies, but 1,066 companies were successfully interviewed. 98. Sectors. The sample was split into four sectors: manufacturing, retail, construction, and other services (hotels and restaurants; whole- salers; transport, storage, and communications; IT services; repair of motor vehicles; and professional, scientific, and technical services that include legal and accounting services, business and consulting serv- ices, architecture and engineering, advertising, research and media companies, private education institutes, and private health clinics). 99. Registration. The sample was split between 600 registered and 400 unregistered companies. Eventually, 647 registered and 419 unreg- istered firms were surveyed. 100. Location. Ten cities were covered for this survey, weighted according to size and the level of industrial activity in the city (table 8). TABLE 8 Sample by City Proposed no. of Actual no. of City businesses surveyed businesses surveyed Kabul 300 306 Herat 125 131 Mazar-e-Sharif 125 142 Jalalabad 125 129 Kandahar 125 121 Ghazni 45 45 Khost 45 71 PuleKhumri 45 48 Kunduz 45 50 Bamyan 20 23 TOTAL 1000 1066 48 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 101. Size. While there were no fixed quotas for firm size, preference was given to the larger companies from the Afghanistan Investment Support Agency (AISA) database. To ensure that micro businesses (those with fewer than five employees) were covered, the proposal specified that these firms would make up a minimum of 5 percent of the sample. 102. Contacting registered companies. The fieldwork team sought registered companies through the following two sources, each of which contained telephone numbers and addresses. This approach enabled the team to target specific businesses that matched the sample criteria: I World Bank panel list (the 338 businesses interviewed in 2005) I AISA list (companies selected by the World Bank, based on size) 103. World Bank panel list. It proved difficult to reach these compa- nies for the following reasons: I After three years, many of the phone numbers had changed. I The address listed was often not specific or was inaccurate. I Twenty of the panel list companies reached by the team refused to answer the questionnaire, having been surveyed not only in 2005 but numerous times since then by various organizations. 104. AISA list. The AISA list also contained inaccurate information, including outdated or incorrect telephone numbers, contact names, and addresses. Again, this made it difficult for the fieldwork team to access the companies on the list. 105. Retailers. It proved challenging to reach sufficient registered companies from the retail sector, partly because there are proportion- ally fewer registered retailers in Afghanistan compared with other sec- tors. Also, retailers are not registered with AISA but with the Ministry of Commerce and Industry, whose privacy policy prevents the distri- bution of company lists. Thus, alternative methods were adopted to fulfil the sample criteria for retailers. 49 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 106. Alternative lists. In addition to the panel list and the AISA list, it was necessary to use other sources to reach companies. Local contacts included relevant nongovernmental organizations (NGOs) (such as the Peace Dividend Trust) and local AISAs, chambers of commerce, and industries. To reach retailers, we used locally obtained Ministry of Commerce lists or, more commonly, local knowledge of large retailers. 107. Contacting unregistered companies. Supervisors were able to contact local organizations--including municipalities, local business associations, unions, and NGOs--to obtain lists of unregistered com- panies that matched the sample quotas. Where this was not possible, the "snowball" technique was used: after a successful interview with an unregistered business, the interviewers asked the respondent to direct them to other unregistered businesses in the city. METHODOLOGY 108. Questionnaire. The questionnaire used the World Bank's core questions as well as some additional questions introduced by the local partner. Altai Consulting translated the questionnaire into the two local languages, Dari and Pashto, and then back-translated into Eng- lish. Three versions of the questionnaire were generated to provide tai- lored tools for the fieldwork team: I The core questionnaire was used for the construction and other services sectors. I The core questionnaire and the manufacturing module were used for manufacturers. I The core questionnaire and the services module were used for retailers. 109. Screening. A screener document ensured that potential respon- dents fit the sample frame. It contained basic questions relating to loca- tion, sector, size, and current operations. The screening was typically con- ducted over the phone, although interviewers made some personal visits because of the inaccuracy of phone numbers provided for specific firms. 50 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 110. Length of interview. Interviews lasted from one to two hours, excluding interruptions. 111. Training.: Training for the fieldwork team was in two phases: (i) one day training to the supervisors by the World Bank (ii) five more days of training by the consulting firm for the full fieldwork team. 112. Pilot interviews in Kabul: Each supervisor and interviewer undertook a pilot interview in Kabul, amounting to 24 pilot interviews in total. 113. Monitoring: This was done at three levels: I The first layer was with local supervisors who reviewed question- naires completed by interviewers, question by question. I The second layer was directly ensured by national and international consultants who reviewed the questionnaires received in Kabul before passing them on for data entry. I The third level was done by World Bank monitors, whose software checks picked up anomalies in the data. 114. Quality control. At the completion of fieldwork, 130 companies were randomly selected and re-contacted to verify that the interview took place in the correct manner and to confirm that accurate infor- mation had been collected. One hundred of these recontacts were con- ducted by the consulting firm; to ensure impartiality, 30 recontacts were undertaken by an independent monitor with no previous involvement in the project. 115. Security. Security was a significant factor in carrying out field- work across Afghanistan, where political unrest and criminal activity present notable risks and obstacles to effective work. Risks were mini- mized in the following ways: I Engaging local interviewers for each city, if possible. Except for Bamyan (in a relatively peaceful part of the country), interviewers operated in their home towns. 51 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 I Flying members of the fieldwork team to training in Kabul, rather than going by roads that are partly controlled by insurgents or ban- dits. This was the case for supervisors and interviewers from Kan- dahar and Herat. I Providing sufficient authority to the interviewers--including an Altai Consulting photo ID, a letter from the Ministry of Commerce and Industry, and a letter from the World Bank--to allay concerns of and potential hostility from prospective interviewees. 116. Incentives. Gifts were offered to respondents to encourage par- ticipation, including a leather-bound notebook, a pen, and a SIM card. A pamphlet in Dari was distributed that explained how the informa- tion from the survey would be used. CHALLENGES AND LIMITATIONS 117. Refusal to interview. Some respondents who had been inter- viewed before expressed impatience at what they considered a waste of time, as well as disillusionment at never having received any tangible benefits from previous interviews. 118. Interview fatigue. Owing to the length of the interview, the qual- ity of the data might have declined toward the end of the questionnaire. 119. Challenge for interviewers. Despite their interviewing experi- ence, the level of business/financial concepts used in the questionnaire proved to be a significant challenge for some interviewers. Interview- ers struggled with concepts such as indirect exports, ATMs and credit cards, and terms relating to the legal status of businesses. Monitors also observed a few errors in recording answers related to costs and rev- enues; the most common error was recording an insufficient number of zeros when the figure was over a million. In some cases, interviewers had trouble calculating percentages and required support from super- visors and consultants in Kabul. All such inaccuracies were corrected; however, a margin of error is to be expected with questions of this type. It is important to acknowledge the difficulties in conducting surveys and carrying out private sector research in Afghanistan. 52 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 INTERVIEWER PROFILE 120. Team size. The fieldwork team included 24 members: 6 supervi- sors and 18 interviewers. 121. Team qualifications. The interviewers were generally college educated in Afghanistan or a neighboring country, and/or had previ- ous experience conducting interviews in the country. Supervisors were more experienced interviewers, typically with a professional back- ground, and acted as managers for the cities they covered. RESPONDENT PROFILE 122. Respondents were owners of the business or the most senior manager working in the establishment--the person considered the "boss" or CEO equivalent. If this person was not available, the second in command was interviewed, provided he or she indicated an ability to answer detailed questions relating to legal status, revenue, and oper- ating practice. 53 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 ANNEX 3. METHODOLOGY FOR PRODUCTIVITY ANALYSIS 123. The 2008 Afghanistan Enterprise Survey dataset includes a rich set of variables relating to inputs and investment climate factors that affect productivity and growth. As noted by Escribano and Guasch (2005), the information collected by the World Bank on investment climate vari- ables is not the kind of information that is routinely observed by econo- metricians.21 This information, therefore, allows controlling for several factors that go unobserved in econometric studies. The productivity dataset brings together survey data on firm-level performance by com- bining and processing key productivity questions from World Bank investment climate surveys carried out in different countries in recent years. Its main purpose is to support comparative productivity analysis. 124. A simple productivity analysis was carried out for the 225 manu- facturing firms in the sample. Firms in the services and construction sectors were not included, because specific questions relating to input usage (value of raw materials and value of machinery, in particular) were designed specifically for the manufacturing sector and were not part of the core survey. 125. An augmented Cobb-Douglas specification was used to estimate, in a single step, the impact of investment climate variables on produc- tivity (Escribano and Guasch 2005). The specification used was: K ln salesi = 0 + 1 ln labi + 2 ln rmi + 3 ln asseti + IC + regional dummies + , k k i k =1 where i indexes the firm, lab represents the number of permanent workers, rm the value of raw materials, and asset the value of assets after depreciation. K IC variables are also included and refer to the investment climate variables that affect productivity. See table 9 below 21 [A. Escribano and J. L. Guasch. (2005). Assessing the impact of the investment climate on pro- ductivity using firm-level data: Methodology and the cases of Guatemala, Honduras, and Nicaragua. The World Bank, Policy Research Working Paper Series: 3621. 54 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 TABLE 9 Variable Definitions for Productivity Analysis Classification Variable Definition Production Function Variables Sales ln(value of total annual sales) - dependent variable Employment ln(total number of permanent workers) Materials ln(annual cost of raw materials and intermediate goods) Assets ln(value of assets after depreciation) Investment Climate and Perception Variables, and General Information Power dummy variable=1 if firm has no electricity Theft dummy variable=1 if firm experienced losses due to theft, robbery, vandalism, or arson Security dummy variable=1 if firm pays for security Interpretation of laws dummy variable=1 if firm finds government officials' interpre- tation of laws and regulations inconsistent and unpredictable Banking dummy variable=1 if firm has a bank account Financing of working capital ln(proportion of working capital financed from retained earnings) Internet access dummy variable=1 if firm uses e-mail or has its own website Mobile phones ln(percentage of employees that own a mobile phone) Import license dummy variable=1 if firm has an import (trade) license Age of firm ln(difference between 2008 and the year in which the firm began operations) Perception of policy dummy variable=1 if firm perceives policy enforcement as enforcement a major or very severe constraint Perception of corruption dummy variable=1 if firm perceives corruption as a major or very severe constraint Firm type dummy Dummy variable=1 if registered firm. Regional dummies These categories indicate the region where firm is located. Regions classified into Kabul, Herat, Kandahar, Mazar-e-Sharif, and Smaller Cities (Ghazni, Baghlan, Kunduz, Bamyan, and Khost). The omitted category is Kabul. Firm size dummies These categories indicate size of the firm in terms of number of employees. The categories are micro (<5), small (5­19), medium (20­99), and large (100 and above). Micro is the omitted category. "ln" in the variable definitions denotes natural logarithm. 55 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 for details of variables used in the regression analysis. The regression was estimated using ordinary least squares with robust standard errors. 126. Productivity gaps expressed in relative percentage terms were obtained from the estimated coefficients on the region dummies using the formula [exp (coefficient of the region dummy) ­ 1 ]100, for any region other than Kabul, which is the base category. Similarly, the impact (in relative percentage terms) that the kth investment cli- mate dummy variable (k=1,2, . . .,K) has on productivity is given by [exp( k)-1]100. 127. This productivity analysis should be viewed only as a prelimi- nary step in understanding the impact of investment climate (IC) vari- ables on firm productivity. Potential sources of bias and inconsistency in the results may arise due to endogeneity of some of the right-hand side variables and failure to account for some important IC variables due to an insufficient number of observations. 128. There could also be measurement errors in this analysis due to limited sample size and possible data quality issues; furthermore, a number of the econometric issues cannot be resolved without panel data. This report has nonetheless attempted a limited productivity analysis within the existing constraints, with the caveat that the find- ings should be viewed as strictly indicative. However, most of our find- ings triangulate well with the survey indicators and with other studies on Afghanistan, so we have a certain degree of confidence in the broad messages emerging from the analysis. 56 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 ANNEX 4. AFGHANISTAN AND COMPARATOR COUNTRIES Afghanistan Yemen Pakistan Ethiopia Tajikistan Thailand DR Congo Country comparisons GDP (PPP) per capita (US$)22 758 2,412 2,739 897 2,019 8,225 328 Doing Business ranking 2009 162 77 98 116 159 13 181 Starting a business 22 77 50 118 168 44 154 Dealing with construction permits 140 93 33 59 178 12 141 Employing workers 30 136 69 95 128 56 175 Registering property 174 97 48 154 46 5 152 Getting credit 178 59 172 123 172 68 163 Protecting investors 181 24 126 113 150 11 150 Paying taxes 49 124 138 37 159 82 153 Trading across borders 179 71 126 152 177 10 160 Enforcing contracts 160 154 41 78 23 25 173 Closing a business 181 53 87 74 97 46 150 Transparency International CPI 2008 176 134 141 126 151 80 171 Electricity Percentage of firms owning generators 75 53 20 - - - - Frequency of power outages (avg. # of 23 183 34 5 9 - - times in a month) Duration of power outages (hrs) 11 3 2 - - - 18 Output losses for those firms that 9 10 10 1 15 - - experienced power losses (% of sales) Percentage of electricity from generator 77 51 6 - - - - Crime Annual cost of security (as % of sales) 3 - 4 1.1 3 0 3 (continued) 22 IMF (2008). 57 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 Afghanistan Yemen Pakistan Ethiopia Tajikistan Thailand DR Congo Corruption Percentage of contract value paid as 6 16 1 - - - 81 informal payment to secure contract Bribes as percentage of total annual 2 9 34 - - - 84 sales to "get things done" Percentage of senior mgt's time spent on 7 16 2 4 12 1 6 govt regulation requirements Percentage of firms saying gift/ payment required for: (a) an operating license 23 62 0 3 39 39 14 (b) tax inspections 26 - 28 4 33 - - (c) import license 29 45 0 - - - - (d) an electricity connection 32 46 5 - - - - (e) a telephone connection 7 31 1 - - - - (f) a construction permit 31 33 1 - - - - Finance Share of working capital from banks 1 9 2 - - - - Percentage of new investments from banks 2 9 - - - 75 3 Percentage of enterprises that have a 51 <5 63 - - - - bank account Percentage of enterprises with a loan 5 17 9 46 34 - - from a financial institution Permits and Licenses Days to obtain operating license 16 11 19 11 23 37 18 Days to obtain construction-related permit 88 18 30 61 63 37 22 Days to obtain import license 13 8 14 14 14 37 18 Innovation and Technology Percentage of firms with internationally 5 19 9 4 17 45 4 recognized quality certification Percentage of firms using technology 6 17 3 4 25 - 4 licensed from foreign companies Percentage of firms with formal training 14 36 5 38 21 - - programs for permanent full-time employees 58 THE AFGHANISTAN INVESTMENT CLIMATE IN 2008 Afghanistan Yemen Pakistan Ethiopia Tajikistan Thailand DR Congo Percentage of Enterprises that: (a) use e-mail 30 - 26 - - - - (b) have own website 15 - 95 18 21 34 6 Trade Average time to clear direct exports 12 8 4 4 20 1 4 through customs (days) Average time to clear imports from 15 14 7 14 15 4 13 customs (days) 59