Addressing Regulatory ‘Software’ Barriers to Business Growth 86243 Chandana Kularatne and Jose Lopez-Calix World Bank Policy Paper Series on Pakistan PK 07/12 December 2012 _______________________________________________ This paper is a product of the South Asia Poverty Reduction and Economic Management Unit (SASEP). It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at Jlopezcalix@worldbank.org. Abstract This policy paper explores the relative importance of the software regulatory barriers to growth in Pakistan. Such software barriers have been identified as part of the major constraint in the “Framework for Economic Growth� of the Government of Pakistan. Indeed, adequate “software� is needed to provide an environment in which the ‘hardware’ of growth (physical infrastructure) could be expanded and made more productive. Among possible ‘software’ constraints, the findings of various international surveys allow to disentangle the relative importance of multiple possible regulatory barriers; first by identifying what is ‘in the books,’ and then by assessing what is actually experienced ‘on the ground’ by entrepreneurs. Following the ensuing prioritization of the identified barriers, this paper suggests that the new growth strategy would benefit from focused policy efforts in seven key areas, where regulatory barriers and perceived obstacles are most constraining to business development: getting electricity, paying taxes, enforcing contracts, registering property, obtaining construction permits, starting a business, trading across barriers, and having access to finance (particularly among small firms). The paper also expounds a detailed description of the provincial disaggregation of those barriers, which attempt to complement the general findings and allow for provincially-led customized solutions. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development / World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Addressing Regulatory ‘Software’ Barriers to Business Growth Chandana Kularatne and Jose Lopez-Calix This paper was motivated from early conversations with Eric Manes and Shahid Javed Burki during their process of dissemination of a private sector strategy for Pakistan, and from discussions with Nadeem Ul Haque and Vaqar Ahmed during the preparation and implementation of the new Framework for Economic Growth. It has benefitted from inputs of Deepak Bhattasali. The paper is an outgrown of work extracted from Doing Business Surveys and Enterprise Surveys. Addressing Regulatory ‘Software’ Barriers to Business Growth Introduction 1. The main challenge of growth in Pakistan, according to the Framework for Economic Growth (FEG)1 is to address the ‘software’ of economic growth. In so doing, the FEG states that Pakistan has more of a software problem rather than a shortage of hardware (physical infrastructure). By software, the FEG refers to broad issues of economic governance, institutions, incentives and human resources that prevent a proper environment in which the ‘hardware’ of growth could be expanded and made more productive at every level. A ccording to FEG, Pakistan has been implementing defective public sector projects, allocating untargeted and arbitrary incentives–subsidy and granting protection to lagging productive activities. This has adversely affected the efficiency of infrastructure development while the system of incentives and protection has stifled the development of a competitive marketplace. In response, the strategy encourages the improvement of productivity by making investment more effective and increasing competitiveness in the market; easing the entry and exit of firms; reforming tax and tariff policy and eliminating distortions; and, limiting government involvement. The strategy argues that entrepreneurship and innovation—which drive productivity and growth—could be greatly encouraged by improving the quality of regulation and, in so doing, governance; reforming and strengthening institutions; promoting market development; and, developing civil service reform. Policy incentives would imply reforming the legal and judicial framework and taxation systems, and reforming restrictive zoning laws, which have impeded the growth of domestic commerce and hampered the role of cities as generators of economic growth. 2. Besides regulatory barriers, the paper prioritizes the existence of economic distortions and poor functioning of domestic markets. Likewise claiming that the cost of doing business in Pakistan is due to excessive regulation, combined with weak institutions and inadequate market development as a constraint to business development. Whereas the current regulatory framework represses domestic commerce (retailing, warehousing, and transport), construction and city development; heavy government direct participation in agriculture, storage, transport, construction, to name a few, is also stifling investment. 3. Among such set of overarching factors that affect the business environment in Pakistan, the degree to which different regulations limit the ability of firms to increase productivity, scale and profitability is diverse and varied. It depends on whether the firm is large or small, formal or informal, the market structure and industry the firm operates in, and firm location. The ability of firms to navigate the regulatory obstacles, bottlenecks in logistics, labor and tax regulation and infrastructure hurdles (and also their impact) varies across firms. 4. Henceforth, this policy paper makes a step forward to sharpen the analysis of the FEG by disentangling the relative importance of different regulatory barriers, per the laws write-up and application itself; and assessing obstacles to market development as perceived by entrepreneurs. To do so, it (i) analyzes the business environment in Pakistan based on the Doing Business (DB) Survey 20122 and investigates how the indicators of doing business in Pakistan change with location of the firm;3 and (ii) assesses the business environment for manufacturing firms in Pakistan across firm size using the Enterprise Surveys (ES) 2007 and 1 Government of Pakistan. 2011. “Framework for Economic Growth: 2011.� Planning Commission. Islamabad. 2 World Bank. Doing Business 2012. http://www.doingbusiness.org/reports/global-reports/doing-business-2012 3 Data on location are from 2010. This caveat is made due to data availability. 1 Addressing Regulatory ‘Software’ Barriers to Business Growth 2010 conducted by the World Bank.4 While Doing Business captures data on the formal regulatory requirements faced by firms, the Enterprise Surveys provide statistics on firm perception of the business environment. By combining the information from both sources, it is possible to disentangle the importance of regulatory barriers and institutional obstacles to market development. 5. The relationship between DB indicators and economic outcomes is well documented in the economic literature. It is strongest, when looking at cross-economy correlations and across sectors within an economy to identify sectors more exposed to certain regulations, both in terms of delays or costs. And even if empirical work cannot prove causation, there is some evidence that the number of procedures and time needed to complete transactions are often longest in low income economies, and that positive changes in DB indicators are associated with higher rates of investment, firm entry and lower unemployment, and that the impact of reforms is somewhat more significant in better-governed economies.5 Evidence from Doing Business Survey on Business Regulations 6. The Doing Business indicators benchmark twelve areas of regulations across time. They are: (i) starting a business; (ii) transferring property; (iii) dealing with construction permits; (iv) accessing credit; (v) employing workers; (vi) enforcing contracts; (vii) protecting investors; (viii) trading across borders; (ix) paying taxes; (x) closing a business; (xi) getting electricity; and, (xii) resolving insolvency. The data counts the number of procedures and the time and costs of compliance, relying on inputs from a small number of professionals in each area. It is important to realize that each indicator reflects the de jure procedures and regulations that firms face and not the de facto restrictions firms experience for these different aspects of business environment. The coverage extends to 183 economies.6 7. Out of 183 economies, Pakistan is ranked 105 in the overall ease of Doing Business (DB) index for 2012. Pakistan does better than most South Asian economies except for Sri Lanka and the Maldives (Figure A1.1, Annexure A1). However, this absolute (level) indicator reflects a marked deterioration from rank 96 the previous year,7 a similar trend to that for three other South Asian nations—Afghanistan, Bangladesh and the Maldives. 8. A relative indicator of Pakistan’s position is the measure of each of its DB indicators’ distance from the benchmark (frontier country).8 Pakistan appears close to the frontier in areas such as protecting investors, dealing with construction permits or registering property, but relatively far from it in areas like getting electricity, resolving insolvency and enforcing contracts. Between 2005 and 2012, getting electricity and trading across borders have deteriorated the most in terms of distance from the frontier. 4 For the Enterprise Surveys 2007 and 2010, the respondents were asked questions regarding FY2006 and FY2009 respectively. See http://www.enterprisesurveys.org/ 5 Bhattasali, Deepak, Mary Hallward-Driemeier and Yue Li. 2009. “The Ease of Doing Business in APEC. The Impact of Regulatory Reforms.�APEC Economic Committee EC03/2008T. Singapore. 6 World Bank. Doing Business 2012. http://www.doingbusiness.org/reports/global-reports/doing-business-2012 7 For both Doing Business Surveys in 2011 and 2012, the total number of countries is 183. 8 In the Doing Business Report, the distance from the frontier measure shows the distance of the Pakistan economy to the (benchmark) frontier country for a particular indicator measuring the ease of doing business. It is based on the most efficient practice or highest score observed for each DB indicator across all economies and years included in the DB sample since 2005. This distance from the frontier measure allows one to assess how much the Pakistan economy’s regulatory environment has changed over time, and how far it has moved toward (or away from) the most efficient practices and strongest regulations in areas covered by the survey. 2 Addressing Regulatory ‘Software’ Barriers to Business Growth 9. In contrast, an absolute indicator of Pakistan’s position indicates that the country scores worst in getting electricity (166) followed by paying taxes (158). On the contrary, the country scores relatively well in protecting investors rights (29) and, to a lesser extent, the ease of getting credit (67) (Figure 1 and Figure A1.2, Annexure A1). The following section examines in more detail Pakistan’s performance with respect to e ach DB indicator in descending order of importance. Figure 1 Distance from the Frontier Source: Doing Business database Getting Electricity 10. Pakistan stands at 166 in the ranking of 183 economies on the ease of getting electricity while the median for South Asia is 118. According to data collected by Doing Business, getting electricity requires 6 stages, takes 206 days and costs 1,346 percent of income per capita. In the region, Pakistan does better than only Afghanistan, Bangladesh and Nepal with respect to cost of electricity.9 This is time consuming—just the final stage of obtaining electricity takes an average of 4 months and includes the completion of external connection works, meter installation, erection of poles, laying of cables, completion of sub- station work and providing connection to the location. 11. There is a wide variation across cities in terms of time taken to get a connection. In 2009, connecting a warehouse to the power system took on average 40 days in Peshawar while it takes up to 5 weeks longer in Hyderabad, Sukkur, and Karachi. Comparing the situation 3 years ago, the process is becoming increasingly time-consuming in a number of cities. For instance, in Quetta, getting power takes an average of 70 days in 2009 while in 9 According to the International Energy Agency’s (IEA) World Energy Outlook 2011, Pakistan has an electrification rate of 62.4 percent in 2009 which falls below the South Asian average of 68.5 percent. See http://www.iea.org/weo/ 3 Addressing Regulatory ‘Software’ Barriers to Business Growth 2006 it took 41 days. The situation has also worsened in Lahore and Faisalabad where the time has increased by 12 and 14 days, respectively, during the same period. Table 1 Getting Electricity Procedures Time Cost Country Rank (number) (days) (% of income pc) Afghanistan 104 4 109 3,957 Bangladesh 182 7 372 3,526 Bhutan 145 6 101 1,265 India 98 7 67 216 Maldives 132 6 101 709 Nepal 99 5 70 1,996 Pakistan 166 6 206 1,346 Sri Lanka 95 4 132 1,192 South Asia Median (2012) 118 6 105 1,306 Source: Doing Business Survey 2012 Ease of Paying Taxes10 12. In Pakistan it is difficult and costly to pay taxes. Globally, Pakistan stands at 158 in the ranking of 183 economies on the ease of paying taxes. When measured by the cost of paying taxes, Pakistan is marginally below the South Asian average of 18.6 percent of profits but above the Organisation of Economic Cooperation and Development (OECD) high- income countries average of 15.4 percent. But once, labor taxes and other taxes, and contributions are accounted for, Pakistan’s cost of paying tax is 35 .3 percent of profits in 2012. Such total tax rate (as a percentage of total profits) has been on a downward trend, falling from 43.6 percent in 2006.11 This is driven by a decrease in the corporate income tax from 55 percent in 1992 to 35 percent by 2012. In addition, small companies pay a discounted corporate income tax rate, amounting to only 20 percent of taxable profits.12 13. The tax burden is unevenly distributed among taxpayers. The Income Tax Ordinance of 2001 contains 70 pages of exemptions which shelter many activities from taxation. The tax burden falls disproportionately on a few sectors, notably manufacturing.13 Furthermore, most taxes are collected from a few large manufacturing firms. The preferential corporate income tax regime for small companies is based on a threshold —PKR 250 million in turnover and 250 workers.14 This encourages firms in Pakistan to remain small and informal. 14. Although the total cost of paying taxes in Pakistan is below the OECD and South Asian average, the compliance cost of taxes weighs heavily on entrepreneurs in Pakistan. Doing Business 2012 reveals that Pakistan does poorly on number of payments per year and 10 The paying taxes indicator records all taxes and mandatory contributions that a medium-size company must pay in a given year. It also measures the administrative burden of paying taxes and contributions. 11 Doing Business excludes value-added taxes from the total tax rate because they do not affect the accounting profits of the business—that is, they are not reflected in the income statement. 12 Doing Business in Pakistan 2012 reports that Pakistan increased the profit tax rate for small firms. 13Alm, James, and Mir Ahmad Khan. 2008. “Assessing Enterprise Taxation and the Investment Climate in Pakistan.� International Center for Public Policy Paper Series, Andrew Young School of Policy Studies, Georgia State University. Paper 810. http://aysps.gsu.edu/isp/files/ispwp0810.pdf 14 Section 153 of the Income Tax Ordinance of Pakistan 2001. See http://www.cbr.gov.pk/ 4 Addressing Regulatory ‘Software’ Barriers to Business Growth the time required to comply with tax requirements. On average, firms make 47 tax payments a year, spend 560 hours a year (about 14 weeks) filing, preparing and paying taxes which is approximately double of the South Asian average. This finding is compounded by the fact that the Federal Bureau of Revenue (FBR) can modify tax legislation ad hoc. For instance, 75 changes were made between 1990 and 2007.15 Continuous change in the rules increases compliance cost and undermines the trust of taxpayers in the fiscal system. As such, in Pakistan, it is more likely that a larger share of economic activity ends up in the informal sector—where businesses pay no taxes at all. In 2007, only 2.3 million taxpayers were registered in a country of 156.7 million people, the equivalent of 1.5 percent of the population.16 15. When it comes to paying taxes, there are not many local variations within Pakistan. Doing Business 2010 (sub-national series) shows that entrepreneurs in all cities spend the same amount of time on the same number of payments (around 560 hours), but for a number of payments that varies little, except for Islamabad. The capital city does not impose social security contributions (Figure A1.3, Annexure A1), and therefore it is less cumbersome to pay taxes in the federal capital (35 payments) than in the other cities of Pakistan (47 payments). Ease of Enforcing Contracts17 16. In Pakistan, the ease of enforcing a contract is hampered by the number of procedures and time. Pakistan ranks 154 out of the 183 economies on the ease of enforcing contracts. Doing Business 2012 finds that enforcing a contract takes on average 46 procedures and 976 days in Pakistan, double the time it does in the OECD —31 procedures and 518 days. The time required to complete procedures has increased by 96 days between 2004 and 2011 in Pakistan. As the procedures to resolve disputes get complex, the less likely firms are to report that judges are impartial and court decisions fair.18 In addition, it costs 23.8 percent of the value of the claim to enforce a contract, 4 percentage points higher than it does in OECD. Countries such as Pakistan that perform poorly in the enforcement of contracts, limit their ability to produce and export customized and advanced products, than those with sound contract enforcement.19 17. The time taken to enforce a contract varies among the different cities. Despite the identical laws and regulations of the court system at the national level, and a number of procedural steps taken to enforce a contract, being the same irrespective of the court’s location within Pakistan,20 the time it takes to enforce a contract varies from city to city. This 15 Thirsk, Wayne. 2008. “Tax Policy in Pakistan: An Assessment of Major Taxes and Options for Reform.� International Center for Public Policy Paper Series, Andrew Young School of Policy Studies, Georgia State University. Paper 808. http://aysps.gsu.edu/isp/files/ispwp0808.pdf 16 Martinez-Vazquez, Jorge. 2006. “Pakistan: A Preliminary Assessment of Federal Tax System.� International Center for Public Policy Paper Series, Andrew Young School of Policy Studies, Georgia State University. Paper 624. http://aysps.gsu.edu/isp/files/ispwp0624.pdf 17 The Ease of Enforcing Contracts indicator measures the efficiency of the judicial system in resolving a commercial dispute. It calculates the time, cost and number of procedures involved from the moment the plaintiff files a lawsuit, through trial and then judgment, until the actual payment through a public sale of the defendant’s movable goods is made. 18 Nunn, Nathan. 2007. “Relationship Specificity, Incomplete Contracts, and the Pattern of Trade.� Quarterly Journal of Economics. 12 (2):569–600. 19 ibid. 20 Resolving of commercial disputes through Pakistani courts is regulated by a single federal law—Code of Civil Procedure, 1908. http://www.ma-law.org.pk/pdflaw/CODE%20OF%20CIVIL%20PROCEDURE%201908.pdf 5 Addressing Regulatory ‘Software’ Barriers to Business Growth is so because the provision of legal services, the governing rules and implementation of directives, court performances and budgets vary depending on location.21  An entrepreneur has to wait for an average of 6 years to enforce a contract in Peshawar—the longest time among all cities. At the other extreme, in Sukkur, where contract enforcement is easiest, an entrepreneur waits for 1,060 days before getting the judgment satisfied. Filing and service of initial court documents in Gujranwala and Rawalpindi takes 40 to 45 days. The process takes longest in Peshawar (90 days) and Karachi (96 days)—3 times longer than in Islamabad and Faisalabad (Figure A1.4, Annexure A1).  The cost to enforce a contract is also cheapest in Sukkur (20.6 percent of the claim value) and most expensive in Lahore (42.8 percent of the value of the claim) where attorney fees are among the highest across the cities measured.22 The fees for filing court proceedings are the second highest cost after attorney fees. 23 Court fees are cheaper across cities in Sindh, Balochistan, and Khyber Pakhtunkhwa provinces, but are more expensive in Punjab (Figure A1.5, Annexure A1).  Across all cities, the time needed to go through a trial and obtain final judgment is the most time consuming. On average, the trial and judgment phase for a contractual dispute takes about 66 percent (840 days) of the total time to enforce the contract. In Islamabad, the trial and judgment takes 1,000 days of the total time of 1,395 days needed to enforce a contract, similar to Dhaka (1,047 days) and Mumbai (1,095 days).24  The time needed to enforce judgment in Pakistan is on average 378 days with a wide variation across cities. For instance in Peshawar it can take up to 900 days to enforce judgment while in Faisalabad it takes 100 days.25 Limited court capacity in cities increases the time taken to enforce a judgment, with varying costs of enforcement across cities. It is cheapest to go through enforcement proceedings in Karachi (1.2 percent of the claim value) and most expensive in Islamabad at (4.6 percent of the claim value).26 This is explained by the higher costs associated with publication in the newspaper and the public sale of the defendant’s movable goods. Ease of Registering Property27 18. Registering property in Pakistan requires 6 procedures, takes 50 days and costs 7.7 percent of the property value. There has been hardly any change in these indicators since 2005. Archaic filing systems, low accountability and work overload combined give rise to 21 Doing Business 2010 finds that if an entrepreneur regularly follows law suit for enforcement of a contract, the judgment can be obtained sooner than if the entrepreneur just waits, which could take 5 years or even longer. 22 The cost in Sukkur is similar to the cost in Canada and New Zealand (22% of the value of the claim), while Lahore is comparable to Kenya (47% of the value of the claim). 23 Filing fees are set by the Court Fees Act in the respective provinces and cities of Pakistan. Court fees include case filing fees, expert fees and other court costs. Expert fees are also set by the court. 24 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 25 Furthermore, in Peshawar there is a more limited market for the public sale of the defendant’s movable goods and as such enforcement of judgment takes longer. 26 Enforcement fees include registering the judgment, organizing the public sale of the goods and other enforcement costs. 27 In the Doing Business Survey the registering property indicator represents procedures, time, and costs necessary to transfer a property title from one business to another, when a company purchases land and a building from another company. Doing Business in Pakistan 2012 assumes that the property is registered and free of title dispute. 6 Addressing Regulatory ‘Software’ Barriers to Business Growth rent-seeking behaviors, and delays adversely affect the registering of property in Pakistan.28 Due to differences in regulations and practices, local variations in procedures, time and cost are observed. The number of procedures does not vary per city. Five procedures are identical across the country: (i) obtain a proof of ownership (called a fard) from the Revenue Office; (ii) purchase a stamp paper, (iii) pay the stamp duty and capital value tax at the bank; (iv) hire a deed writer to draft the sale deed; and (v) present the document to the registration office for final registration. Only in the province of Sindh are entrepreneurs required to publish the property transaction in a newspaper to invite objections. In some provinces— Punjab, Khyber Pakhtunkhwa and Balochistan—entrepreneurs must personally follow up on the transfer of property title with the patwari.29 19. Transferring the property title is the most time-consuming procedure, accounting on average for 71 percent of the total time needed to register property in Punjab. The fastest city is Lahore (30 days) and the slowest Quetta (52 days).30 These differences stem from varying requirements: (i) Cities in Sindh (Sukkur, Karachi and Hyderabad) require individuals to have a newspaper publication informing people of the transfer of property, adding an additional 8 to 14 days; and (ii) The Revenue Office’s efficiency in transferring the property title varies depending on the degree of computerization and workload. 20. The cost of registering a property in Pakistan is substantially higher than the median cost in South Asia, 7.7 percent as compared to 5.9 percent of the property value, respectively.31 A property buyer must pay four types of fees, collected by different levels of government: (i) a provincial stamp duty, (ii) registration fee, (iii) a municipal Urban Immovable Property Tax (UIPT) levied by the Town Municipal Authority, and (iv) a federal capital value tax. In 2011, Pakistan made registering property more expensive by doubling the capital value tax to 4 percent (Figures A1.6, Annexure A1). Although all, but one, taxes/fees are the same in all provinces—local variations arise from the stamp duty rates set at the provincial level: 2 percent of the property value in Punjab, 3 percent in Sindh and Khyber Pakhtunkhwa and 5 percent in Balochistan. Islamabad is an outlier, as it does not enforce the UIPT. As a result, it is cheapest to register property in Islamabad (7 percent of the property value) and most expensive in Quetta (11 percent of the property value) (See Figure A1.7, Annexure A1). 21. There have been reforms instituted at both the local level such as the computerization of land records in Lahore and Sialkot in the Punjab. Chosen as a pilot city for the government of Punjab, Lahore had computerized the revenue records of 57,000 acres of land belonging to 60,000 land owners by 2003. This reform improves the efficiency of Lahore’s registrars and patwaris such that from 2006 to 2009, the time to register a deed and transfer a property title in Lahore decreased by 3 and 4 days, respectively. A pilot program to computerize deed registration is also underway in Sialkot, which reduced registration time from 13 days in 2006 to 6 days in 2009.32 28 Usman Qazi, Muhammed. 2006. “Computerization of Land Records in Pakistan.� UNDP Asia-Pacific Development Information Program, A Case Study. 29 The patwari is the local official at the Revenue Office in charge of issuing the fard and completing the property transfer. 30 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 31 When Maldives is excluded as an outlier, the median for South Asia is 5.1 percent of the property value. 32 Staff Report. 2003. “Data of City Land Records Completed.� Daily Times. December 15, 2003. 7 Addressing Regulatory ‘Software’ Barriers to Business Growth Dealing with Construction Permits33 22. More cumbersome and costly procedures are associated with an increased likelihood of informal payments in exchange for construction permits. In Pakistan, rigid, archaic rules in the issuing of construction permits push the construction sector into the informal economy and highlight the paucity of effective “software� in dealing with construction permits. The relatively poor performance of Pakistan in dealing with construction permits results in 1 entrepreneur out of 4 expecting to pay facilitation fees to public officials in order to get things done.34 By some estimates 60 to 80 percent of construction projects in developing economies are undertaken without a building permit, because the approval process is too complex or oversight too lax.35 23. According to Doing Business 2012, dealing with construction permits requires 11 procedures, takes 222 days and costs 262.5 percent of income per capita in Pakistan. Globally, Pakistan stands at 104 in the ranking of 183 economies on the ease of dealing with construction permits. The median for South Asia is 123 days. The number of procedures, time and cost involved vary significantly across cities since licenses for site development, building permits and utility connections are governed at the local level. 36 Expanding businesses that need to build a warehouse would find it easier to obtain all approvals and utility connections in Peshawar, Gujranwala and Lahore than in Hyderabad and Quetta. When it comes to the time taken, from pre-construction procedures to utility connections, it takes 124 days in Peshawar but almost twice as long in Karachi. For instance, in Karachi, it takes 2 months to obtain water and sewerage connection and 2 months to receive a completion certificate from Karachi Building Control Authority (KBCA). In most cities, hooking up a newly built warehouse to the water, sewerage and electricity networks, and obtaining a fixed telephone line takes at least 2 months. In Peshawar, Multan, Sialkot and Faisalabad the process is usually completed 20 days faster (Figure A1.8, Annexure A1). 24. The cost of obtaining construction-related permits also varies by city. For instance, Doing Business 2010 reports that in Sheikhupura, the administrative cost of building a warehouse is 380.3 percent of income per capita while in Islamabad, it is 797.9 percent of income per capita. The variations between cities stems from differences in fees for building permits and certificates of completion. All cities charge building permit fees, depending on the use and size of the building. In Hyderabad the development authority charges 6 different fees to process and validate the building plan as opposed to 1 or 2 fees charged in other cities. Although, after construction is completed, the certificate issued after final inspection is free in most cities, authorities in Gujranwala, Hyderabad, and Islamabad charge fees between PKR 5,000 and PKR 70,000.37 33 Dealing with construction permits measures the procedures, time, and cost required for a business in the construction industry to obtain the necessary approvals to build a commercial warehouse and connect it to electricity, water, sewerage and telecommunication services. 34 World Bank. Enterprise Survey. 2007. http://www.enterprisesurveys.org 35 Moullier, Thomas. 2009. “Reforming Building Permits: Why Is It Important and What Can IFC Really Do?� International Finance Corporation. The World Bank Group. Washington D.C. 36 Moreover, it is not just the number of procedures but the actual number of days multiplied by the number of procedures which determine the ease of obtaining construction permits. 37 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 8 Addressing Regulatory ‘Software’ Barriers to Business Growth Starting a Business38 25. Research shows that there is a direct relationship between the number of procedures to register a business and informality—fewer procedures to start a business are associated with a smaller informal sector.39 According to data collected by Doing Business 2012, starting a business in Pakistan requires 10 procedures, takes 21 days, costs 11.2 percent of income per capita and requires paid-in minimum capital of 0 percent of income per capita. That is 3 days less than the South Asian median (25 days), and less than in India (29 days). The cost is below the South Asian median of 18.5 percent of income per capita. However, the number of procedures (10) is higher than the median for the region (7). The time in days to register a business has remained relatively static; only marginally reducing from 24 days to 21 days, while in South Asia, the regional average has declined from 33 days to 26 days. 26. Despite the same regulatory framework across the country, there are differences in registration time due to varying efficiency of local branches, variations in the use and availability of online or official services, and the type of transaction involved. Business startup time varies from 16 days in Islamabad to 24 days in Gujranwala. Company name verification and incorporation takes 6 days in cities with a high usage of e-Services and between 7 to 10 days in cities where the vast majority of entrepreneurs continue to submit documents in person.40 Also, since there are no Security and Exchange Commission (SECP) offices in Gujranwala, Sheikhupura, Sialkot, Hyderabad and Rawalpindi, entrepreneurs must travel to Lahore, Karachi, or Islamabad to register their companies, adding up to 1 day and travel costs of PKR 80–700. 27. Major variations affect the time involved in applying for a national tax number and a sales tax number at the tax facilitation center of the Regional Tax Office (RTO) of the Federal Board of Revenue. Registering for income tax takes 2–7 days, with the longest time recorded in Sheikhupura as it does not have an RTO office and therefore entrepreneurs must travel to Lahore to register.41 Registration for sales tax takes on average 6 days across the 13 cities. The process is relatively fast in Karachi, Islamabad and Rawalpindi (1 to 4 days) and slow in Sheikhupura (10 days). The main sources of delays are the lengthy verification process for all applications and insufficient coordination between the local RTOs and the Central Registration Office in Islamabad. Finally, it takes less time for firms to register their employees for social security in Punjab than everywhere else. As part of the Punjab Industrial Policy 2003, the authority to register new enterprises with the Punjab Employees Social Security Institution was transferred from provincial headquarters to the local district level. The revision of documents became faster and the time to register for social security fell by 4 days in Faisalabad and Lahore as a result. 38 The starting a business indicator measures all procedures that are officially required to start up and formally operate a commercial or industrial small- or medium-size limited liability company. These include obtaining necessary licenses and permits and completing required notifications, verifications and registrations for the company and its employees with the relevant authorities. 39 Djankov, Simeon, Rafael La Porta, Florencio López-de Silanes, and Andrei Shleifer. 2002. “The Regulation of Entry.� The Quarterly Journal of Economics. 117 (1):1–37. Cambridge, MA: MIT Press 40 In September 2008, the SECP and e-Government Directorate launched e-Services—a new web portal for business incorporation. The e-Services site enables entrepreneurs to check availability of the proposed company name and reserve it, download all necessary forms and incorporate their new businesses online. Companies registering through e-Services no longer need to obtain a company seal; instead the applicants can obtain a digital signature from the National Institute of Facilitation Technologies. 41 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 9 Addressing Regulatory ‘Software’ Barriers to Business Growth 28. In terms of the cost of business registration, Pakistan has reduced it from 40 percent of income per capita in 2004 to a low of 5.8 percent in 2010. 42 In order to promote online registration, the fees are lower for this option. There are also additional costs when entrepreneurs must travel to another city to complete their registrations, as is the case in Sialkot. Trading Across Borders43 29. Regulations that affect the time and cost spent on paperwork, clearance procedures, port activities and inland transportation are just another form of “software� that may hinder commerce between countries. In Pakistan, exporting a standard container of goods requires 7 documents takes 21 days and costs US$660. This is a considerable decline from 31 days and $996 in 2006. Importing the same container of goods requires 8 documents, takes 18 days and costs US$705. Whereas the time taken to import has declined since 2006 by 21 days, the cost to import has increased by US$388. In so doing, Pakistan outperforms South Asia in terms of number of documents needed to trade, time and cost required for such trade. Indeed, the average cost to import/export for South Asia is over double that of Pakistan, with nearly twice the number of procedures as that in Pakistan. Moreover, it is cheaper for a firm in Pakistan to import/export goods than the average of high-income OECD countries. However, when it comes to the number of documents and time taken to import/export goods, Pakistan falls short of the OECD average. 30. Despite significant variations in the distance of each city from the port of exit and entry, the time required to import and export a standard 20-foot container does not vary significantly among 13 main cities. On average, the export process takes 22 days, while the import procedures take 20 days. It is easiest to import and export from Karachi, where the country’s major ports are located—22 days to export and 18 days to import—followed by Hyderabad and Sukkur. Exporting and importing a container takes the longest from Quetta. (Figure A1.9, Annexure A1). However, distance determines substantial differences in the cost of trading across all cities. It costs 57 percent more to import a container to Lahore (US$ 1,088) than to Quetta (US$ 693).44 Karachi is the cheapest city in which to conduct trading activities. Moreover, a study calculates the loss from export delays in Pakistan at around 1 percent of trade for each extra day. For perishable agricultural products exported by Pakistan, the cost is nearly 3 percent of the volume of trade for each day’s delay45 (Figure A1.10, Annexure A1). Ease of Resolving Insolvency 31. Fast and cheap insolvency proceedings facilitate the speedy return of businesses to normal operation in periods of financial stress. A well-functioning insolvency system can facilitate access to finance, save more viable businesses and thereby improve growth and 42 The cost to complete a procedure includes only official costs, no bribes and no professional fees 43 The trading across borders indicator measures all procedural requirements and the associated time and cost (excluding trade tariffs) for exporting and importing a standard cargo of goods by ocean transport for which payment is made by letter of credit. 44 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 45 Djankov, Simeon, Caroline Freund, and Cong S. Pham. 2010. “Trading on Time.� Review of Economics and Statistics. 92(1):166–73. 10 Addressing Regulatory ‘Software’ Barriers to Business Growth sustainability in the overall economy. Pakistan’s rank on resolving insolvency has fallen from 71 to 74 out of 183 countries between 2011 and 2012, but this improvement is marginal as Pakistan has kept constant the time required to resolve insolvency (4 years) and the associated cost (4 percent of the value of the estate). However, except for Maldives and Sri Lanka, Pakistan still outperforms most South Asian countries in terms of this indicator. Ease of Getting Credit46 32. Pakistan scores better than most countries in South Asia for the ease of getting credit—it stands at 67 in the ranking of 183 economies on the ease of getting credit—but is situated in a region which fairs poorly with respect to the ease of getting credit. However, Pakistan lags the South Asian regional average and the OECD average in terms of the private bureau coverage—only 2 percent of individuals and firms listed in the largest private credit bureau are covered, whereas it is 5.8 percent in South Asia and 64 percent in the OECD. The economy also has a score of 4 on the depth of credit information index (81 st rank) and a score of 6 on the strength of legal rights index (73rd rank).47 These indicators have remained relatively constant since 2004. Protecting Investors 33. Pakistan does relatively well on the protection of investors at 29 in the ranking of 183 economies. The country also does well on measures related to extent of disclosure, extent of director liability and ease of shareholder suits. Except for Bangladesh, Pakistan does better than all other South Asian countries. Unfortunately, the Doing Business survey does not investigate whether or not these statutes are enforced. In Pakistan, as in other countries, these statutes may not be enforced as consistently as one would hope. Thus even though rules and regulations exist to protect investors, potential investors may shy away as they may recognize that these rules could exist only on paper. Findings from Enterprise Surveys on the Business Environment48 34. Doing Business Surveys and Enterprise Surveys can be used together to see if economies with fewer delays or lower costs contribute to better firm performance and vice versa. Enforcement of those regulations plays a critical part in how these constraints are perceived by entrepreneurs. Studies already indicate that the general pattern observed is that more procedures and longer compliance times are indeed associated with more time spent by managers dealing with officials.49 These patterns stand, in general terms, for Pakistan. Indeed, whereas the Doing Business Survey was analyzed to determine the areas where Pakistan falls short in terms of regulations that affect businesses, this section considers 46 Doing Business assesses the sharing of credit information and the legal rights of borrowers and lenders with respect to secured transactions through the depth of credit information index, which measures rules and practices affecting the coverage, scope and accessibility of credit information available through a public credit registry or a private credit bureau; and the legal rights index, measuring the degree to which collateral and bankruptcy laws protect the rights of borrowers and lenders. 47 The strength of legal rights index is between 0 to 10—higher the index, greater the protection of rights of borrowers and lenders through collateral laws and bankruptcy laws. The depth of credit index is between 0 to 6 – higher the index, greater the scope and accessibility of credit information distributed by public credit registries and private credit bureaus. 48 This section highlights key findings of the Enterprise Surveys for Pakistan (2007 and 2010), including only those firms that have majority private ownership. 49 Bhattasali, Deepak, Mary Hallward-Driemeier and Yue Li. 2009. “The Ease of Doing Business in APEC. The Impact of Regulatory Reforms.�APEC Economic Committee EC03/2008T. Singapore. 11 Addressing Regulatory ‘Software’ Barriers to Business Growth firms’ perceptions regarding how effective those potential barriers are felt as an obstacle to business development. Overview of the General Obstacles Faced by Firms 35. Between 2007 and 2010, electricity, crime, theft and disorder, political instability, practices of competitors in the informal sector and business licensing and permit were the constraints perceived by firms to firm development (Figure A2.1, Annexure A2). 36. In absolute terms, whereas in 2007 most firms ranked Customs and Trade Regulations as the main obstacle; in 2010 the main constraint was Getting Electricity. The significant decrease in Customs and Trade regulations is consistent with the perceived rankings (and improvements) in trading across border of the Doing Business. Instead, approximately 60 percent of firms claim that electricity supply is their biggest obstacle in 2010. About 98 percent of firms cited experiencing power outages over the year as opposed to 90 percent during 2007. This may indicate that the electricity disruptions have become a binding constraint to firm activity and the problem (Figure A2.2, Annexure A2). The limited and irregular supply of electricity is followed by Political Instability and Crime, Theft and Disorder in order of importance.50 37. Access to finance and taxation issues appear as decreasing, but still important, second-rank obstacles. About 50 percent of firms surveyed by the Enterprise Survey 2010 reported access to finance as not an obstacle versus 35 percent by the Enterprise Survey 2007. Still important, this finding indicates a slight improvement in access to finance in Pakistan over the period, even after limited bank financing was available following the 2007- 08 global financial crises.51 Interestingly enough, this finding is consistent with Bhattasali et al.52 which point at electricity and access to finance as top constraints perceived by entrepreneurs in low income economies. Whereas tax administration—jointly perceived with macro and political instability and corruption, as part of the second set of constraints— becomes less relevant an obstacle in the Enterprise Survey 2010. Tax rates become the 8th (out of 13) more relevant obstacle. An interesting caveat pointed out by World Bank53 is that economic governance issues—corruption and taxation—are more relevant than infrastructure or finance in the services sector, contrary to manufacturing and agriculture. 38. In Pakistan, the obstacles firms face may have a disproportionate affect on small vis- à-vis large firms and, consequently, on firm productivity. Disentangling which obstacles tend to have a higher adverse impact on small vis-à-vis large firms (and vice versa) is also crucial to understanding the degree to which “software regulations� affect business development. By using Enterprise Survey data (2007 and 2010)54, this section shifts the focus to perceived constraints across firm size for the manufacturing sector.55 In particular, the next section addresses two questions: 50 Across 2007 and 2010 survey rounds, only access to finance, corruption, electricity, labor regulation and transport were commonly tested for severity of the obstacle in both surveys. 51 One of the implications of the global financial crisis in 2008-2009 was to limit bank financing to firms. 52 Bhattasali, Deepak, Mary Hallward-Driemeier and Yue Li. 2009. “The Ease of Doing Business in APEC. The Impact of Regulatory Reforms.�APEC Economic Committee EC03/2008T. Singapore. 53 The World Bank. 2010. “Pakistan Investment Climate: Laying the Foundation for Renewed Growth.� Report No.46435-PK. Washington D.C. 54 Depending on the availability of data, either one or both surveys will be used to conduct the analysis. 55 The set of manufacturing firms analyzed in this study comprise joint ventures with foreign private individuals or owned solely by either citizens or foreigners. 12 Addressing Regulatory ‘Software’ Barriers to Business Growth 1. What are the current constraints that majority of small firms consider as being an obstacle to current operations of the firm?56 2. What are the dimensions of the business environment that encourage small firms to remain small? This question focuses on those aspects that a majority of large firms respond as being a severe or major obstacle to their current operations.57 Obstacles to Production as Perceived by Small Firms 39. There are two important characteristics of the manufacturing sector in Pakistan. The first one is that the manufacturing sector is dominated by small firms employing less than 5 employees. The second one is the fact that irrespective of firm size, approximately 70-80 percent of the manufacturing firms are over 10 years-old, which indicates that there are barriers to entry of new firms. Obstacles may have a relatively more adverse effect on small rather than on large firms’ productivity. Electricity 40. Relative to country comparators, firms in Pakistan suffer the most severe financial losses due to power interruptions, with small firms bearing the highest losses. 58 The Global Competitiveness Report 2011-1259 ranks Pakistan 126 out of 142 countries for the overall quality of electricity supply. The Enterprise Survey (2007) shows that across all firm size categories, electricity proved to be the severest obstacle to firms—60 percent of small firms versus 36 percent of large firms. Thus, the degree to which firms perceived these constraints affecting their business productivity varies according to their size. For instance, the Enterprise Survey (2010) reveals that over 50 percent of firms of each firm size (small and medium) cite electricity as a severe obstacle while 52 percent of large firms claim electricity is not an obstacle to their functioning.60 The reason for this discrepancy may rest with the fact that large firms are more able to circumvent the limited and irregular supply of electricity than smaller firms. Even firms with access to electricity face high costs and highly unreliable supply of power with frequent outages. The power sector in particular, has proved to be a major bottleneck for the economy and a serious threat to overall competitiveness (Figure A2.3, Annexure A2). 41. Electricity gaps have a differential impact between small and large firms. Lack of reliable supply of electricity is less of a constraint to large firms vis-à-vis small firms with 73 percent of large firms owing a generator as compared with only 14 percent of small firms. Thus nearly 40 percent of large firms claim that power outages had no impact on sales while 56 Note that an obstacle may not be perceived to be an obstacle if it is bypassed. For example, courts may be bypassed if informal arrangements are dominant in a country with inefficient judicial institutions; generators may also be used for power supply when electricity supply fails, etc. Therefore, an obstacle may be perceived as less binding if some firms are able to sidestep it. 57 Doing Business in Pakistan 2007 (sub-national series) states Pakistani entrepreneurs may choose not to register and remain informal to avoid excessive regulation that they face as their size increases. This may help to explain the existence of a large number of small firms in the manufacturing sector. 58 The FEG states that the energy crisis is not primarily due to capacity gaps, but to management and organizational difficulties that have resulted in cash flow constraints. It says that the energy sector is characterized by a lack of institutional strength to undertake effective integrated planning, policy development and implementation. 59 World Economic Forum. “The Global Competitiveness Report 2011 -12.� Geneva Switzerland 2011 http://www3.weforum.org/docs/WEF_GCR_Report_2011-12.pdf 60 The firms examined are classified as Small (5-19 employees), Medium (20-99 employees) and Large (above 99 employees). 13 Addressing Regulatory ‘Software’ Barriers to Business Growth only 20 percent of small firms make this claim. Moreover, the loss as a percent of total annual sales due to power outages for the average small firm is 5 percent while for the average large firm is 10 percent. Furthermore, small firms experience more power outages than large firms in 2010. The number of power outages in 2010 for small firms increased to 120 from 20 in 2007 (Figure A2.4, Annexure A2). The median number of power outages in 2010 increased for all firms while the median duration of each power outage declined from 2007 to 2010. There are numerous reasons why small firms tend to experience more power outages than large firms such as location (rural vs. urban), being on an industrial park or not, etc. 42. Moreover, with rising fuel costs, smaller firms are unable to afford the operating costs of running their own generator, which also affects their productivity. Among those firms that own a generator, a higher percentage of larger firms claim that over 50 percent of their power needs are met from own-generators. Taking into account all of the above, the Enterprise Survey 2007 reveals that although small and medium enterprises are the engines of employment creation for the manufacturing sector in Pakistan,61 the average small firm is less productive than the large firm.62 Finance 43. There is an abysmal gap between small and large firms when it comes to access to finance. Barely 27 percent of small firms claim that access to finance is not an obstacle to current operations compared with 50 percent of large firms. Small firms finance 65 percent of all their working capital and 90 percent of all their purchases of fixed assets through retained earnings. In contrast, only 41 percent of large firms finance all their working capital through retained earnings and 61 percent of small firms finance all their purchases of fixed assets through retained earnings (Figure A2.5, Annexure A2). The reasons are multiple:  The lack of an independent assessment of the financial status of small firms.63 An auditor’s report is a standard fiduciary requirement for large firms, but not for small firms, which creates an information asymmetry. Since small firms usually do not have such reports, banks tend to limit their exposure to small firms.  Small firms do not have the infrastructure to apply for a loan.  Large firms have more short-term access to liquidity choices. Although 74 percent of small firms do have a checking and/or savings account,64 only 9 percent of small firms have an overdraft facility while nearly 65 percent of large firms have such a facility. The existence of such facility increases the probability of a firm being successful when applying for a loan, but its lack reveals that banks refuse to provide it —only 7 percent of small firms in 2006 had an overdraft line of credit vis-à-vis 46 percent of large firms.  Land is a major asset used as collateral across all firm size—93 percent of small firms and 84 percent of large firms—but SMEs are much less likely to use movable property as security, compared to large firms. One reason for this lies in the restrictive policy that only allows firms registered in the companies’ registry to secure loans using movable property—putting SMEs at a disadvantage. Approximately, 50 percent of small firms also use machinery as collateral. 61 See Annexure 2 which describes the structure of the Pakistani firms. 62 Annexure 3 examines productivity in the manufacturing sector in Pakistan across firm size. 63 Only 8 percent of small firms have had their annual financial statement checked and certified by an external auditor. This stands in stark contrast to 82 percent for large firms. 64 95 percent of large firms have checking and/or savings account. 14 Addressing Regulatory ‘Software’ Barriers to Business Growth  Smaller firms bear tougher conditions when accessing finance—a higher median collateral-to-loan ratio of 140 percent versus 130 percent for large firms and a lower value of land owned, even after controlling for loan size. Smaller firms own relatively less valuable land. Very high collateral requirements limit loan amounts and exclude those firms that may not have sufficient collateral to fulfill their financing contracts.65 This explains the reason for the high proportion of small firms that use retained earnings to finance either working capital or purchase of fixed assets.  Even though a large percentage of small firms own land, weak property rights disproportionately and adversely affect their ability to use their land as collateral. The process of obtaining titles appears to be grounded in informal payments, particularly for small Pakistani firms. As a result, financial intermediaries are more risk averse towards using land as collateral from small than from large firms. Transport 44. The average percentage value of the consignment lost due to breakage or spoilage is higher for small firms than large firms. The average consignment value lost to breakage and spoilage is 13 percent for small firms and 7 percent for large firms when transporting within the country. All goods needed to be transported by small firms are more likely to get transported using the firm’s own means of transportation. 60 percent of small firms transported 100 percent of their goods using their own transport vis-à-vis 40 percent of large firms. This indicates that either the mode of transport available to small firms is poor or the localities to which small firms transport goods has poor road infrastructure which increases the likelihood of damage to goods transported. Technology 45. Not only do small firms have limited access to cutting edge technology, they may not need such technology as the goods they produce are at the lower end of the technology spectrum. If we use technology from a foreign-owned company as a proxy for access to cutting edge technology, the Business Enterprise Survey (2007) finds that less than 1 percent of small firms use such technology unlike 24 percent of large firms. Hence, if small firms were to increase their productivity and move up, the absorption of such technology may be a necessary prerequisite. However, small firms have limited capability to absorb new technology. The majority (59 percent) of production workers in small firms only have between 0-3 years of educational attainment. With such a low skill base it is unlikely that these firms will be able to innovate or be in a position to absorb new technology that would increase their productivity through technological enhancement (Figure A2.6, Annexure A2). 65The value of collateral relative to the loan size, is higher in Pakistan (134.5 percent) compared to India (129.7 percent), Sri Lanka (106 percent) and Bangladesh (94.6 percent)―as well as Philippines, Malaysia, Thailand, Chile, Brazil and Egypt. 15 Addressing Regulatory ‘Software’ Barriers to Business Growth If Small Firms are Handicapped, Why They Choose to Remain Small? 46. There are many rational reasons why firms in Pakistan may intend to stay small. They apply to the taxation, labor regulation and other regulatory burdens. Taxation and Labor Regulations 47. A majority of firms in Pakistan (65 percent) find the application of regulations as inconsistent and unpredictable and by remaining “informal� and consequently small, firms avoid the constraints imposed on them by over-regulation. In Pakistan, the informal sector could be as large as 35-40 percent of the official economy. Given that large firms cater to a different market than small firms, they are unaffected if small firms are able to avoid the regulatory burdens. 48. Small firms may choose to remain out from the radar of the authorities in order to avoid a variety of regulatory burdens.66 Due to poor enforcement of registration requirements by authorities, firms are able to circumvent registration and avoid hassles. For instance, senior management in large firms spends an average of 13 percent of their time dealing with regulations, while management in small firms only spends 5 percent. 49. Small firms are less likely to be visited by tax inspectors. For instance, during 2005- 06 the average number of times small firms were inspected by a tax official or required to meet with tax officials is 2.7 times. In contrast, large firms had to meet with tax officials 6.7 times.67 When comparing the same firms in 2007 and 2010, the data shows that senior management in small firms spends less time on regulation than large firms. In 2010, both average number of inspections by tax officials and average time spent on regulation by management increased in 2010 from 2007. This implies increasing regulatory burden for this set of firms68 (Figure A2.7, Annexure A2). 50. For Pakistan, labor regulations do not appear to be a significant obstacle to firms — especially small. About 71 percent of small firms do not find labor market regulations to be an obstacle to current operations in contrast to 47 percent of large firms. In addition, Pakistan fairs badly when compared to other developing economies with respect to both the difficulty of hiring and firing of workers and the cost of hiring and firing workers. 69 However, only 6 percent of small firms claim that labor market regulations affected their decision to hire and fire permanent workers while 26 percent large firms assert that these regulations affect their turnover of labor (Figure A2.8, Annexure A2). It may be inferred from all this evidence that labor market regulations, though rigid, are not enforced and/or firms are able to circumvent these regulations. There are two reasons for this: (i) firstly, small firms are less likely to be audited by labor inspectors (only 45 percent of small firms report either being inspected, unlike 78 percent of large firms), and (ii) secondly, on average a large firm is inspected 4.2 times while a small firm is inspected 3.2 times per year. 66 The procedures for registering a business usually apply to firms with more than 10 employees. 67 Note that these are small firms with more than 10 employees which are legally required to register for tax purposes. 68 Doing Business 2007 claims that Pakistani companies remain small and informal, given that the preferential corporate income tax regime for small companies is based on an abrupt threshold —PKR 250 million in turnover and 250 workers. 69 The World Bank. 2010. “Pakistan Investment Climate: Laying the Foundation for Renewed Growth.� Report No.46435-PK. Washington D.C. 16 Addressing Regulatory ‘Software’ Barriers to Business Growth 51. The impact of over-regulation and weak institutions in Pakistan leads to a high incidence of informal payments either for services or for avoiding regulations by all firms. Approximately 75 percent of firms state that it is common to pay an informal payment/gift to get things done when dealing with regulatory impediments to their business operations. However, a higher percentage of large firms (70 percent) know what the exact value of this payment/gift is, unlike only 60 percent of small firms. 52. A final word is devoted to regional differences in the perception of relevant obstacles. Only results for the Enterprise Survey 2007 are available. Beyond electricity which is unanimously considered as a major constraint, Baluchistan sees corruption as the most important issue; whereas Punjab considers macroeconomic stability as a top constraint, and to a lesser extent Sindh; barely 10 percent of firms see political instability as a concern in NWFP, not so with tax rates, customs and trade, and courts. Conclusions and Policy Directions 53. The FEG strategy concentrates on improving the “software� of economic growth. The analysis presented in this paper attempts to prioritize among some elements of the software—in particular those regulatory barriers across the various indicators of the business environment and the perceptions of the obstacles entrepreneurs face on which the country should focus its efforts. The discussion below suggests broad policy directions on which the Government can embark on to improve the business environment. 54. As getting electricity comes at the top of obstacles in both diagnostics, reducing load- shedding should be the top priority for business development. The solution should combine software elements addressing cash and institutional shortcomings and solving the need for integrated planning, policy development and implementation with hardware elements involving selective infrastructure investments to expand supply. 55. As second priority, which is also at the top of both diagnostics, the country should focus in reducing the burden of taxation regulations. This implies addressing both taxation uncertainty, when it refers to tax rates (as indicated in the Enterprise Surveys) as well as efforts in facilitating tax payments, while reducing the time and cost involved in such procedures. More specifically, this would require from Federal Bureau of Revenue (FBR) a mix of (a) simplification of the tax system by eliminating the many exemptions and preferential treatments that riddle the tax system and erode fiscal revenues; (b) expansion of the electronic filing and payment systems to reduce the transaction costs involved with paying taxes; (c) increase the number of regional tax offices across the country so as to facilitate not only tax payments, but business registration; (d) multiply effective mechanisms for firms and individual taxpayers to notify about corrupt officials, so as to prevent informal payments; and (e) conduct frequent taxpayer satisfaction surveys so as to identify new ways to improve FBR services to the client. 56. As third priority there is a need to improve procedures to enforce contracts. A mix of actions should combine a thorough streamlining of filing and court procedures, so as to reduce the number and time devoted to obtain a resolution; creation of additional courts — especially for commercial disputes; expansion of court capacities, both in terms of infrastructure as well as human resources; and reduction of legal (and court) fees. 17 Addressing Regulatory ‘Software’ Barriers to Business Growth 57. The fourth priority pertains to registration of property. In this regard, reforms are needed to simplify and consolidate the procedures for land registration and transfer by computerizing land records and holding patwaris accountable to all cities in Pakistan. There is a need to improve the efficiency of the revenue office through computerization, greater accountability and enforcing strict sanctions by imposing a time limit to settle and obtain the final property title; and to reduce the number of fees and introduce a flat fee for stamp duty so that individuals do not have an incentive to undervalue their property. Furthermore with a flat fee, registrars and patwaris would not need to spend as much time verifying the property value. This would speed up the registration process and significantly reduce the time needed to register property. 58. The fifth priority should ease procedures for obtaining construction permits. Although there have been some recent reforms to ease the regulations when dealing with construction permits70 eliminating overlapping agencies administering bylaws and presiding over inspections and approvals will help to achieve the desired facilitation. In addition inspections need to be more efficient and uniform, either by moving from system of random inspections to a system of risk-based inspections, where inspections take place at critical phases of the construction process; and providing training to staff so that they understand the rules and regulations and avoid administering them in an ad-hoc manner. 59. The sixth priority should ease procedures for starting a business. In order to do this, the Government could increase the availability of online registration services across Pakistan, especially those addressing tax registration procedures. However, in Pakistan, where only 10.5 percent of the population currently has access to the Internet, using e- services to register a business might have limited impact. This would imply making internet stations available, where entrepreneurs could go to register more cheaply and efficiently. 60. The seventh priority would be to ease trading across borders. There have been limited improvements in reducing the time taken to import goods through Karachi between 2006 and 2010 from 39 to 18 days. The following reforms are still needed to improve the efficiency and reduce the cost of trade both within and between Pakistan and other nations.71 Further steps including reducing and streamlining documentation requirements, strengthening inland clearance facilities, improving the electronic data interchange system and inland transportation especially Pakistan Railways, which suffers from an aging infrastructure. 70 At the provincial level, there have been some reforms when dealing with construction permits. Lahore, Multan, Rawalpindi, Gujranwala, Faisalabad and Sialkot adopted uniform building and zoning regulations that clarify the permitting process and specify required documents. Hyderabad plans to enforce more detailed building and town-planning regulations, while authorities in Quetta are working on new bylaws to replace those of 1937. 71 World Bank. Doing Business. 2010 (sub-national series). http://www.doingbusiness.org/reports/global- reports/doing-business-2010 18 Addressing Regulatory ‘Software’ Barriers to Business Growth Annexure A1 Evidence from the Doing Business Survey based on measures of the Business Environment Figure A1.1 Ease of Doing Business – Country Comparison Afghanistan Bhutan India 2012 Bangladesh 2011 Nepal Pakistan Sri Lanka Maldives 1 51 101 151 Source: Doing Business 2011 and 2012 Figure A1.2 Pakistan: Ease of Doing Business Indicators Protecting Investors 29 Getting Credit 67 Resolving Insolvency 74 Trading Across Borders 75 Starting a Business 90 Dealing with Construction Permits 104 Ease of Doing Business 105 Registering Property 125 Enforcing Contracts 154 Paying Taxes 158 Getting Electricity 166 Source: Doing Business 2012 19 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A1.3 Provincial Comparison of “Payment of Tax� Indicators Social security Corporate income tax Federal other* Pension contributions Provincial taxes contributions Payments Hours Islamabad Capital Territory 35 559 Balochistan 47 560 47 560 Punjab 47 560 Kyber Pakhtunkhwa 47 560 Sindh 0 10 20 30 *Pakistan’s sales tax is not included in the total tax rate, as it does not impact the company’s commercial profits. Source: Doing Business 2010 (Sub-National Series) Figure A1.4 Pakistan: Filing, Judgment and Enforcement of Contract Filing Judgement Enforcement Faisalabad Lahore Gujranwala Karachi Sukkur Sheikhupura Average Sialkot Islamabad Hyderbad Multan Quetta Rawalpindi Peshawar 0 500 1000 1500 2000 Source: Doing Business 2010 (Sub-National Series) 20 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A1.5 Pakistan: Comparison of Various Fees Attorney fees Court fees Enforcement fees Sukkur Peshawar Multan Karachi Hyderabad Faisalabad Rawalpindi Sheikhupura Average Quetta Gujranwala Islamabad Sialkot Lahore 0 5 10 15 20 25 30 35 40 45 Source: Doing Business 2010 (Sub-National Series) Figure A1.6 Cost of Registering a Property (% of Property Value) Nepal 5 Bhutan 5 Afghanistan 5 Sri Lanka 5.1 South Asia Median (2012) 5.85 Bangladesh 6.6 India 7.3 Pakistan 7.7 Maldives 16.7 Source: Doing Business in Pakistan 2012 21 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A1.7 Pakistan: City-wise Comparison of Cost of Registering Property Stamp duty Registration fee Capital value tax Urban immovable property tax Islamabad Sialkot Faisalabad Sheikhupura Lahore Gujranwala Rawalpindi Multan Average Peshawar Sukkur Karachi Hyderabad Quetta 0 2 4 6 8 10 12 Source: Doing Business 2010 (Sub-National Series) Figure A1.8 Pakistan: Time Taken to Obtain Construction Permits Pre-construction During construction Post-construction Utilities Peshawar Lahore Multan Islamabad Gujranwala Rawalpindi Sukkur Faisalabad Sialkot Average Sheikhupura Hyderabad Quetta Karachi 0 50 100 150 200 250 Source: Doing Business 2010 (Sub-National Series) 22 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A1.9 Pakistan: City-wise Comparison of Cost of Registering Property Time to import Time to export Sukkur Karachi Hyderabad Sheikhupura Average Sialkot Peshawar Multan Faisalabad Rawalpindi Lahore Islamabad Gujranwala Quetta 0 5 10 15 20 25 Source: Doing Business 2010 (Sub-National Series) Figure A1.10 Pakistan: Cost-to-Import and Cost-to-Export Cost to import Cost to export Karachi Quetta Hyderabad Faisalabad Sukkur Sialkot Peshawar Average Multan Islamabad Gujranwala Rawalpindi Sheikhupura Lahore 0 200 400 600 800 1000 1200 Source: Doing Business 2010 (Sub-National Series) 23 Addressing Regulatory ‘Software’ Barriers to Business Growth Annexure A2 Evidence from Enterprise Surveys based on Measures of the Business Environment across firm size Figure A2.1 Ranking of Obstacles 14 13 13 2007 12 12 12 2010 11 11 10 10 10 9 9 Greater obstacle 8 8 8 7 7 6 6 6 5 5 4 4 4 3 3 2 2 2 1 1 0 Source: Enterprise Surveys 2007 and 2010 Figure A2.2 Severity of Obstacles 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 2007 2010 2007 2010 2007 2010 2007 2010 2007 2010 Access to Finance Corruption Electricity Labor Regulations Transport No Minor Moderate Major Severe Source: Enterprise Surveys 2007 and 2010 24 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A2.3 Intensity of Severity 100% Transportion 90% Tax rates Tax administration 80% Practices of competitors in informal sector 70% Political instability Macroeconomic policy 60% Labor regulations Inadequately educated labor 50% Functioning of the courts 40% Electricity Customs and Trade Regulation 30% Crime, theft & disorder 20% Corruption Business licencing and permits 10% Access to Land 0% Access to Finance 0.55171281 Small Medium Large Source: Enterprise Survey 2007 Figure A2.4 Median Number of Power Outages 140 Median number of power outages Median duation of power outages (hrs) 2.5 120 2 100 80 1.5 60 1 40 0.5 20 0 0 2007 2010 2007 2010 2007 2010 Small Medium Large Source: Enterprise Survey 2007 25 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A2.5 Access to Finance as an Obstacle No Obstacle Minor obs Moderate Major or Severe obstacle 14 29 24 13 19 22 24 26 22 50 27 31 Small Medium Large Source: Enterprise Survey 2007 Figure A2.6 Education Level in Small, Medium and Large Firms 0 to 3 4 to 6 7 to 12 Above 13 0 3 1 3 17 39 34 35 32 59 47 31 Small Medium Large Source: Enterprise Survey 2007 26 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A2.7 Time Spent on Regulations & Tax Officials Average senior management's time spent on dealing with regulations (%) 16 Average number of times inspected by tax officials 7 14 6 12 5 10 4 8 3 6 2 4 2 1 0 0 2007 2010 2007 2010 2007 2010 Small Medium Large Source: Enterprise Surveys 2007 and 2010 Figure A2.8 Severity of Obstacle for Labor Regulations No obstacle Minor obs Moderate Major obs Very severe obstacle 1 3 2 6 6 8 9 7 17 13 26 26 71 60 47 Small Medium Large Source: Enterprise Survey 2007 27 Addressing Regulatory ‘Software’ Barriers to Business Growth Annexure A3 Characteristics by Firm Size of the Manufacturing Sector The Census of Economic Establishments (2005) provides interesting insight into the manufacturing sector. Out of a total of 583,329 establishments, manufacturing of textile wearing apparel and leather industries account for 252,111 economic establishments (43.22 percent), followed by manufacturing of food, beverages and tobacco at 121,875 establishments (20.09 percent). Interestingly, the number of establishments in the manufacturing of fabricated metal products, machinery and equipment are 58,476 and basic metal industries are 4,511. In terms of numbers, only 10.80 percent are manufacturing enterprises that support other industries as input where significant value addition can take place. Following is a sub-sectoral provincial distribution of manufacturing establishments. It is interesting to note that despite Sindh housing 17.82 percent of all economic establishments, NWFP has a greater number of economic establishments in the manufacturing sector i.e. 15.93 percent (Table A3.1). Table A3.1 Number of establishments in the Manufacturing Sector by Industry Federal Manufacturing Industry Division Total Punjab Sindh NWFP Balochistan Capital Manufacturing of Food, Beverages and 121,875 70,329 14,652 35,030 1,413 451 Tobacco Textile, Wearing Apparel & Leather 252,111 167,141 43,331 35,863 4,684 1,092 Industries Manufacturing of Wood & Wood 63,087 46,766 5,182 10,045 890 204 Products including Furniture Manufacturing of Paper & Paper 10,141 6,291 2,912 761 101 76 Products, Printing & Publishing Manufacturing of Chemicals and Chemical, Petroleum, Coal, Rubber & 5,781 4,170 1,185 362 22 42 Plastic Products Manufacturing of Non-Metallic Mineral 15,476 11,755 2,093 1,516 46 66 Products except Petroleum & Coal Basic Metal Industries 4,511 3,456 496 504 21 34 Manufacturing of Fabricated Metal 58,476 46,850 6,631 4,092 599 304 Products, Machinery & Equipment Other Manufacturing Industries & 51,871 42,461 4,386 4,802 175 47 Handicrafts Total 583,329 399,219 80,868 92,975 7,951 2,316 Source: Census of Economic Establishments 2005 The small and medium enterprises are the engine of employment creation in Pakistan in the manufacturing sector. In the manufacturing sector 91 percent of firms have an employee base of 1-5 persons. 6.6 percent employ 6-10 persons, over 2 percent have 11-50 employees, 28 Addressing Regulatory ‘Software’ Barriers to Business Growth and only 0.2 percent of firms have over 50 workers.72 Therefore to encourage employment growth through private sector development, it is important to increase the productivity of SMEs by improving the investment climate indicators that affect these firms. In fact, improving private sector development is one of the strategies highlighted in the FEG (Figure A3.2). Figure A3.2 Percentage of Small and Medium Enterprises by Employment Size 94.46 4.60 0.89 0.035 0.0087 0.0026 0.0017 1-5 6-10 11-50 51-100 101-150 151-200 201-250 Number of employees Source: Small and Medium Enterprises Development Authority (SMEDA) 72 The 2005 Census of Establishments finds that 94.46% economic establishments have an employee base of 1-5. The employment pattern of these Establishments is such that of a total of 2.95 million, 2.78 million Units (94.46%) employed 1-5 persons, 0.14 million Units (4.6%) employed 6-10 persons, 0.02 million (0.89%) 11-50 persons, and 15,840 (0.048%) over 50 persons. 29 Addressing Regulatory ‘Software’ Barriers to Business Growth Annexure A4 Manufacturing Firm productivity in Pakistan across firm size Productivity in smaller firms is lower than in larger firms. The median labor productivity for the large firms is nearly 200 percent higher than small firms. Given that the manufacturing landscape in Pakistan is dominated by small firms and these firms are, on average, less productive than larger firms, this limits Pakistan’s manufacturing growth potential (Figure A4.1). Figure A4.1 Median Labor Productivity Index Median labour productivity index % difference wrt informal sector 400 300 350 250 300 200 250 200 150 150 100 100 50 50 0 0 Small Medium Large Source: Enterprise Survey 2007 The difference in productivity across firm size cannot be attributed to the fact the large firms tend to be older and therefore have established systems that make them more efficient. The Enterprise Survey 2007 finds that over 50 percent of firms, across all three firm size in the manufacturing sector, are over 10 years old. Across firm size, firms older than 10 years tend to be more productive than those between 1-5 years old. Given that the analysis finds smaller firms less productive than larger firms and that most firms across all sizes are older than 10 years, age does not appear to be the differentiating factor accounting for variation in productivity across firm size (Figure A4.2). Firms located in an industrial zone have higher labor productivity. This may be attributed to productivity gains from agglomeration economies. The greatest impact on productivity by being located in an industrial estate is on medium firms with large firms gaining the least. This may be because large firms that are not in an industrial zone may be able to take advantage of some of the agglomeration effects that accrue to firms which are located in an industrial park (Figure A4.3). Size and type of product matter in general. For a given industry, low productivity small firms co-exist with higher productivity large firms. In the non-metallic mineral industry large firms have the highest relative productivity while in the leather industry there is not much difference in labor productivity across firm size. Moreover, the most productive industry for small firms is the leather industry which incidentally, along with the garment industry, is the 30 Addressing Regulatory ‘Software’ Barriers to Business Growth least productive industry for large firms according to the Enterprise Survey 2007 (Figure A4.4). Figure A4.2 Firm Age <=1 1 to 5 5 to 10 10 to 20 >20 47% 45% 57% 34% 38% 32% 11% 12% 7% 8% 4% 4% Small Medium Large Source: Enterprise Survey 2007 Figure A4.3 Location of Firms Large Medium Small 0 20 40 60 80 100 120 140 Not in Industrial Zone In Industrial Zone Source: Enterprise Survey 2007 31 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A4.4 Size and Type of Product affect Median Labor Productivity (logs) Large Medium Small Other manufacturing Leather Non-metallic minerals Chemicals Machinery & equipment Textiles Garments Food Source: Enterprise Survey 2007 Why do Less Productive Small Firms Exist? This subsection explores why less productive small firms exist within the same industry as larger, more productive firms. The answer to this question will reveal that small firms cater to the local market while large firms tend to cater to the national and international markets. Small firms (50 percent) cater to their surrounding locality while only 14 percent of large firms cater to the locality in which they are situated. Moreover, the term “local market� may be interpreted very differently across firms of different size. For small firms the local market is likely to be their neighborhood while for a large firm it is usually their city. Thus one may infer that even within the locality, the market that large firms cater to is of a larger radius than that of smaller firms (Figure A4.5). For a given industry, there is limited linkage between small and large firms. Small manufacturing firms mainly sell their product to individuals (50 percent) and other small firms (31 percent). Only 4 percent of large firms purchase from small firms. This indicates that small and large firms either produce different products or the same product that is highly differentiated and that there is limited vertical integration between small and large firms. Therefore, as small and large firms in the same industry service different markets, even though the productivity of large firms is greater than that of small firms, small firms may continue to produce at a lower of efficiency (Figure A4.6). The fact that smaller firms are less likely to be able to compete on quality in foreign markets compared to large firms indicates that the products the smaller firms produce are of inferior quality to that of large firms. Only 11 percent of small firms claim that product quality standards are harder to meet in foreign markets unlike only 3 percent of large firms. This supports the theory that not only are products the small firms sell are very different to that of 32 Addressing Regulatory ‘Software’ Barriers to Business Growth large firms but they are of a lower quality and therefore for different markets even within the domestic market73 (Figure A4.7). Smaller firms are also less likely to compete with imports on either price or cost of production. Two findings support this: Firstly, 31 percent of small firms face no competition from foreign imports. In contrast, only 16 percent of large firms face no competition from foreign goods. This result generally holds across all industries. Secondly, a higher percentage age of large firms (49 percent) find foreign competitors having an important impact on production costs of exiting products (Figure A4.8) much more than small firms (24 percent). Therefore, the existence of relatively more productive large firms coexisting with less productive small firms within a particular industry is only possible if the two categories of firms produce differentiated products which cater to different markets. The smaller tend to produce lower quality goods which are sold to their neighborhood market while large firms produce higher quality goods which may be intended for export. Considering who the principal buyers are for each firm-type, the evidence indicates that there is also limited interaction between small and large firms. Thus there is a limited positive externality accruing to small firms from the existence of more productive large firms. The lack of interaction further intensifies the bifurcation of a particular industry and more generally, the entire economy, between these two classes of firms. Figure A4.5 Markets Catered by Different Firm Sizes Local National International 9% 21% 32% 41% 47% 54% 50% 33% 14% Small Medium Large Source: Enterprise Survey 2007 73 Small firms do not enter export markets not because they cannot compete but they lack foreign market information to allow them to determine the characteristics of the export market and the avenues for entry into the foreign market. 33 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A4.6 Markets Catered by Different Firm Sizes Associated company Large firm Medium firm Small firm Individual Government Others 5% 4% 7% 1 1 2 23% 40% 56% 18% 24% 19% 24% 19% 28% 9% 11% 4% 3% Small Medium Large Source: Enterprise Survey 2007 Figure A4.7 Pressure on Production Costs Due to Foreign Competition Does not Not at all Slightyly Important Very Important 14% 18% 33% 10% 13% 13% 17% 16% 33% 19% 33% 22% 30% 18% 10% Small Medium Large Source: Enterprise Survey 2007 34 Addressing Regulatory ‘Software’ Barriers to Business Growth Figure A4.8 Impact of Foreign Competition on Proportion of Firm Size None Limited Moderate Intense 13% 14% 23% 12% 8% 27% 20% 24% 21% 51% 51% 36% Small Medium Large Source: Enterprise Survey 2007 35