MACROECONOMICS, TRADE AND INVESTMENT MACROECONOMICS, TRADE AND INVESTMENT EQUITABLE GROWTH, FINANCE & INSTITUTIONS INSIGHT Gender and Taxpayer Study in Khyber Pakhtunkhwa, Pakistan IN PREPARATION FOR THE ONE STOP TAXPAYER FACILITATION CENTERS Hitomi Komatsu and Irum Touqeer June 2023 © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Attribution—Please cite the work as follows: Komatsu, Hitomi, and Irum Touqeer. 2023. “Gender and Taxpayer Study in Khyber Pakhtunkhwa, Pakistan: In Preparation for the One-Stop Taxpayer Facilitation Centers.” EFI Insight-Macroeconomics, Trade and Investment. Washington, DC: World Bank. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank. org. Cover photo by: Waseem Niaz, PakStockPhoto GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 2 >>> Acknowledgments This study was conducted by the Integrating Gender Equality into Tax Reform project team of the World Bank’s Global Tax Program, which is led by Ceren Ozer (Senior Economist). Hitomi Komatsu (Gender Economist / ET Consultant) in the Fiscal Policy and Growth Unit in the Macroeconomics, Trade & Investment Practice, and Irum Touqeer (Public Sector Specialist, revenue mobilization) in the Macro, Trade, Fiscal and Public Sector Team in Pakistan led the study and authored this report. The authors acknowledge the valuable guidance and advice on the survey design, questionnaire, and implementation provided by Caren Grown (former Senior Advisor), Raymond Muhula (Senior Public Sector Specialist), Mishka Zaman (Senior Social Development Specialist), Isis Gaddis (Senior Economist), Kevin McGee (Senior Economist), Matt Collin (Economist), Emcet Tas (Senior Social Development Specialist), Nick Orgeira Pillai (Consultant), and Hannelore Niesten (Consultant) of the World Bank Group, and participants in the World Bank Global Tax Program Gender Equality and Taxation Seminar Series. The authors extend thanks to Dong Kyu Kwak (Senior Land Administration Specialist), who generously provided detailed information on the Land Records Management and Information Systems (LRMIS) in Punjab province, and Caren Grown, Ceren Ozer (Senior Economist), Moritz Meyer (Senior Economist), Uzma Quresh (Senior Social Development Specialist), Zuzana Boehmova (Consultant), and Gayatri Koolwal (Consultant), who peer reviewed the report. The team would like to express its gratitude for the discussions and collaboration with the senior officials of the Government of Khyber Pakhtunkhwa (KP), representing the Board of Revenue (BOR), Excise, Taxation, and Narcotics Control Department (ETNCD), KP Revenue Authority (KPRA), KP Finance Department, and KP Information and Technology Board, in particular, Mr. Syed Zafar Ali Shah (former Senior Member BOR), Mr. Syed Haider Iqbal (former Secretary ETNCD), Mr. Zafar ul Islam (Director General ETNCD), Mr. Saidul Amin (former Director, ETNCD), Mr. Ahmed Shabbir (Deputy Director Land Records, BOR), and Mr. Faisal Shahzad (PD, KP Revenue Mobilization and Public Resource Management Program; KPRMP). The team is also grateful to the field staff of BOR and ETNCD who shared data, and coordinated and supported the interview logistics with the survey team. The team would like to thank Dr. Rashid Bajwa (CEO, NRSP), Mr. Zahoor Hussain Khan (President & CEO, NRSP Bank), and the heads of their offices in Peshawar, Mardan, and Abbottabad for sharing data and collaborating with the study. The team gratefully acknowledges the women and men who shared their views during the focus group discussions and key informant interviews and in the taxpayer survey. The interviews, discussions, and survey were conducted with the support of the team from Gallup Pakistan, who worked tirelessly to contribute to the design of survey and facilitated the data collection. Key members of Gallup team included Bilal Gilani, Babar Aziz, Amnah Imtiaz, Laila Waqar, and Nayema Siddiqui. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 3 >>> Contents Acknowledgments 3 Executive Summary 6 1. Introduction 9 2. Provincial Tax Offices and Ongoing Reforms in KP Province 12 2.1. BOR: Service delivery centers (SDCs) 12 2.2. ETNCD: Excise and taxation field offices providing taxpayer facilitation services 13 3. Data and Methods 14 3.1. Data collection for the taxpayer survey 14 3.2. Focus group discussions and key informant interviews 16 4. Facilitating Women’s Voluntary Tax Compliance by Understanding Women’s Constraints and Experiences 17 5. Understanding the Context: Women’s Constraints and Experiences 19 5.1. Women’s mobility and access to and use of digital technologies 19 5.2. Gender roles and methods of tax payment and property registration 22 5.3. Gender gaps in access to tax information 24 5.4. Satisfaction with services and challenges at provincial tax offices 26 5.5. Gender differences in trust in KP tax administration and reciprocity 29 GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 4 6. Measures to improve women’s tax payments and registration 31 6.1. Availability of female tax officials and establishment of dedicated female counters 31 6.2. Safe and gender-equitable work environment 32 6.3. Better facilities—washrooms and separate areas for women 32 6.4. Simplification of processes and clearer guidance 33 6.5. Tax officials who speak the local language 33 6.6. Use of digitized services—electronic registration and tax payments 33 6.7. Tax education and information campaigns on women’s property rights 34 6.8. Compilation, digitization, and sex-disaggregation of taxpayer and service user data 34 6.9. Physical access 35 7. Conclusion 37 References 39 Appendix 41 Appendix A: Provincial Tax Revenue, KP Province 41 Appendix B: Survey Sampling Strategy 42 Appendix C: Survey Methodology 43 Appendix D: Number of Study Participants 44 Appendix E: Individual Characteristics of Respondents 45 Appendix F: Reasons for Not Purchasing Goods Online 48 Appendix G: Respondents’ Probability of Paying Provincial Taxes (conditional mean) 49 Appendix H: Taxpayers’ Methods Used for Provincial Tax Payment or Property Registration 50 Appendix I: Use of Social Media as a Source of Tax Information by Age 51 Appendix J: Suggested Timeline of Recommended Action Plan 52 GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 5 >>> Executive Summary The Khyber Pakhtunkhwa (KP) provincial government is undertaking reforms to bring three tax agencies under one roof as one-stop taxpayer facilitation centers (TFCs) across the province to assist taxpayers with property registration, tax filing, and payment. The KP government is also digitizing land records and tax services so that taxpayers can access property records and register property more easily and make tax payments elec- tronically. In conjunction with this effort, the KP government is embarking on an ambitious electronic governance initiative, including the provision of digital public services. Through these reforms, the KP government aims to improve revenue mobilization by increasing the number of taxpayers and facilitating voluntary taxpayer compliance.    The KP government is committed to increasing women’s access to registration and tax services and facilitating women’s tax voluntary compliance, as manifested by hiring, for the first time, 12 female patwaris (field officers), who are responsible for maintaining land records, as well as female inspectors and property tax collectors. Two of the KP tax administrations—the Board of Revenue (BOR) and the Excise, Taxation, and Narcotics Control Department (ETNCD)—are also considering establishing service counters dedicated to provid- ing services to female property owners and taxpayers.     This study’s objective is to support the KP provincial government’s efforts to facilitate women’s voluntary tax compliance by understanding women’s constraints and experi- ences in tax payment and property registration and the gender gaps in access to and use of digital technologies in the context of the ongoing reforms. The study draws data from a survey that interviewed 1,200 current taxpayers and nontaxpayers (a third of whom were women), focus group discussions with women and female tax administrators, and key informant interviews with senior directors of the KP tax administrations. Provincial taxes and services provided at provincial tax offices of BOR and ETNCD, namely motor vehicle taxes, urban im- movable property tax (UIPT), land record (FARD) fees, and land transaction (or mutation) fees are the focus of the study. The digitalization of the KP government’s property registration and tax services presents new opportunities and challenges for women. Land records stipulate ownership and are used for tax purposes, while the UIPT records are often used to record property ownership in urban areas. Centralization and digitization of these records can help improve the security of property rights because they become more accessible and less vulnerable to manipulation by local officials (Khan, Khan, and Zahir 2020). Digitizing services may facilitate women’s access to GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 6 property records, registration of property, and tax payments. It en and men in the ongoing reforms. Identifying the context may reduce taxpayer compliance costs by making it unneces- around women’s low property ownership, women’s restricted sary to travel to registration offices or payment points and in- mobility, gender roles in registration and tax payments, gen- teract face-to-face with government officials. Reduced travel der gaps in access to tax information, access to and use of would be particularly useful for women, who may be limited digital technologies, and trust and reciprocity can help tailor in their mobility due to childcare responsibilities or who may taxpayer programs to support women’s property registration need male chaperones or permission to travel. Yet digitization and tax payments, which in turn can help the KP government could exacerbate gender gaps in accessing registration and increase own-source revenue. Facilitating women’s property tax services because women may lack access to and use of registration may also result in improved property rights, which digital technologies. could lead to other benefits for women, such as generating and diversifying incomes, providing better financial security Understanding women’s constraints and experiences in in cases of emergencies, or serving as collateral for access property registration and tax payments and the gender to finance (Deere and Doss 2006; Meinzen-Dick et al. 2011; gaps in digital technologies can improve the ability of the Pradhan, Meinzen-Dick, and Theis 2019). KP government to address the different needs of wom- Key Findings The study identifies specific challenges women face in 4. Lack of information and unclear processes for prop- property registration and tax payments. It also reveals that erty registration are the most-cited challenges for women’s experiences and challenges are not monolithic, but female users of field tax offices, affecting a higher tend to vary by level of education. proportion of women than men. Lack of female staff is the second most-cited difficulty affecting female taxpayers’ experience. 1. Women’s lack of mobility is likely to restrict their ac- cess to property registration and tax services. Wom- 5. Lack of bathrooms for women is a key concern for fe- en are largely confined to their homes and are less able male tax officials, some of whom have to use bathroom to move freely without permission or male chaperones to facilities at nearby fast-food restaurants. visit places outside their homes, including government of- fices—this applies even more to women with limited edu- 6. There are gender gaps in how women and men access cation (primary schooling or less). tax information, highlighting the need for tailored tax education and information strategies. Women, par- 2. Women with limited education may not fully benefit ticularly less educated women, tend to rely on personal from digitization of tax services. Digital access and use networks—friends and family—to obtain information on vary by education; women with limited education are less taxes. Social media is a more important source of infor- likely than educated women to own mobile phones and to mation for men than for women, particularly for more edu- use digital banking and wallet applications. There are also cated and younger men. gender differences in the use of digital financial technolo- gies; a higher share of men than women use mobile bank- 7. While there is no gender difference in the level of ing applications. Face-to-face service provision, such trust in the provincial tax administration, women are as in one-stop tax facilitation centers in all districts, less likely than men to believe that the government is therefore continues to be important—at least in the using tax revenues to provide government services medium term, until more women and men have better to its citizens. If women do not believe the government is digital technology access and use. using its resources effectively, they may be less inclined to register their properties and pay taxes. 3. Payment of taxes using banking or mobile applica- tions is not yet common, and it is much less common among women than men. Taxpayers currently travel to payment points, such as a bank branch, to pay taxes in person rather than use electronic methods of payment. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 7 Key Recommendations 1. Availability of female tax officials—including female as TV or radio programs and involvement of NGOs or patwaris, tehsildars,1 and tax and excise officials— field staff in disseminating information on women’s prop- and establishing service counters dedicated to serv- erty rights and taxes are important to reach less educated ing women are likely to facilitate women’s property women and men. Different social media channels can be ownership and voluntary tax compliance. But these used to reach out to younger and more educated taxpay- reforms require a review of the institutional structure and ers. Sensitization campaigns can target both women as roles, sustained training, and a safe and equitable work well as men who could play a role in assisting women’s environment—including equitable promotions, flexible property registration. The provincial tax administration work arrangements, anti-harassment policies and prac- websites—which are mostly available only in English— tices, and physical safety—to recruit and retain women. are currently not a major source of tax information; they might be used much more if the information were also 2. Establishing separate bathrooms and eating and pray- given in Urdu or Pashto.   ing areas for women is crucial for female staff and for improving services to female taxpayers. It is recom- 5. Review of the implications of computerization of mended that new infrastructure investments planned for land records, digitization of vehicle registration, and establishing one-stop TFCs include washrooms and ar- establishment of registration process service coun- eas for women. ters for women are important. Women in focus group discussions also emphasized the need for using easy-to- 3. A differentiated strategy for digital access and com- understand terminology in Urdu or Pashto in the guid- munication for women is important to facilitate their ance on tax payments and registration forms, as the use of digital services. These include providing tax- existing guidance was difficult to understand.  payer education and support to women, simplifying digital applications and content, receiving regular feedback from 6. Compiling, digitizing, and sex-disaggregating taxpay- taxpayers on the applications, and ensuring applications er and service user data could help revenue adminis- are available in local languages. trations develop tailored taxpayer programs and ser- vices that meet the needs of both women and men. It 4. Gender gaps in how women and men access tax in- could also help the KP government to determine which formation underscore the need for tailoring tax cam- one-stop TFCs could establish service counters dedi- paigns or information strategies to reach different cated to solely to serving women and direct human taxpayers. Field-level education campaign efforts, such and financial resources for this purpose. 1. The tehsildar is a tax officer who collects land revenue and other fees from landowners in a tehsil (a subdivision of a district) (Ali and Nasir 2010). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 8 1. >>> Introduction Improving tax facilitation and strengthening enforcement measures have comprised the two important pillars of reform programs on tax compliance in low- and middle-income countries (Dom et al. 2022). Facilitating taxpayer compliance typically involves simplifying reporting re- quirements, providing taxpayer support, reducing in-person interactions with tax officials, and providing easier methods of tax payments, such as through banks, online, or using mobile ap- plications (Dom et al. 2022). Enforcement measures include expanded assessments and audits and reducing scope for evasion (Dom et al. 2022). The Government of Khyber Pakhtunkhwa (KP) is undertaking several reforms to increase own-source revenue by facilitating voluntary taxpayer compliance and increasing the number of taxpayers, among other efforts (World Bank 2019b).2 One such reform includes the KP government’s plans to establish one-stop taxpayer facilitation centers (TFCs) that would (a) assist taxpayers with property registration and tax payment, and (b) coordinate among differ- ent provincial tax agencies under one roof (World Bank 2019b). Twelve one-stop TFCs will be established, where the public can access services by the province’s three tax agencies, namely the Board of Revenue (BOR; also called the Revenue and Estate Department); the Excise, Taxation, and Narcotics Control Department (ETNCD); and the KP Revenue Authority (KPRA). In addition, the KP government is digitizing land records, automating and simplifying processes, and establishing common digital platforms to allow taxpayers to register property and make tax payments electronically (World Bank 2021).3 The Government of KP is also embarking on an ambitious electronic governance initiative that includes provision of digital public services, such as digital issuance of marriage, birth, death, and education certificates, in accordance with the KP Digital Policy 2018–2023 (World Bank 2022b). All three reforms are supported by an ongo- ing USD$118 million project funded by the International Development Association (IDA) and the Khyber Pakhtunkhwa Revenue Mobilization and Public Resource Management Program (KPRMP) (World Bank 2019b). 2. Other ways of increasing own-source revenue include expanding the range of taxable activities and assets, enhancing capacity building for tax collection, and increasing nontax revenues (World Bank 2019b). 3. The KP government plans to allow motor vehicle owners to register their vehicles online, while land registration will contin- ue to require registration at BOR offices, based on key informant interviews with DG ETNCD and DD BOR. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 9 This study aims to support the KP provincial govern- records can help improve the security of property rights be- ment’s efforts to facilitate women’s voluntary tax compli- cause they become more accessible and less vulnerable to ance by understanding women’s constraints and experi- manipulation by local officials (Khan, Khan, and Zahir 2020). ences in property registration and tax payments and the Digitization of services may enable women to access property gender gaps in access to and use of digital technologies records and register or pay taxes more easily (World Bank in the context of establishing one-stop tax facilitation centers 2022b). It can reduce taxpayers’ compliance costs by making and digitization of services. It discusses measures that could it unnecessary to visit registration offices or payment points (a) facilitate women’s property registration for land, resi- and have face-to-face interactions with government officials dential, or nonresidential properties and motor vehicles, and (Dom et al. 2022). This would be particularly useful for women (b) make it easier to pay taxes. The study uses both a quanti- who may need their family’s permission or a male chaperone tative approach—through a survey of 1,200 current taxpayers for travel or who have childcare responsibilities that limit their and nontaxpayers, a third of whom are women—and a qualita- mobility. As an example, digitization of land records and wom- tive approach, consisting of focus group discussions (FGDs) en-focused services in the neighboring Punjab province has with women and female staff, and key informant interviews helped increase the number of women with their names on with senior directors of the Government of KP. Provincial taxes land titles (World Bank 2022a). Despite these opportunities, and services provided at provincial tax offices, namely the ur- digitization could also exacerbate the gender gaps in access ban immovable property tax (UIPT), motor vehicle registration to tax services because women tend to lack access to and and token tax,4 land record fee (FARD fee), and land mutation use of digital technologies (World Bank 2022b). The reasons fee, are included in this study.5 for the digital gender gaps could include women’s low literacy, numeracy, or digital skills; unaffordable devices and connec- Women in KP province largely lack ownership or control tivity costs; lack of identification documents required for digi- over property. Only 2 percent of women in KP province are tal IDs; lack of appropriate products and content; and family landowners, versus 28 percent of men, and only 3 percent gatekeeping that restricts women’s digital access (GPFI 2020; of women are residential property owners, compared to 72 GSMA 2021; World Bank 2021; World Bank 2022b). Without percent of men (NIPS and ICF 2019). A myriad of factors con- understanding women’s constraints and experiences with dig- tributes to the gender gap in property ownership and rights, ital services, women could be inadvertently excluded from the including social norms about ownership and inheritance, lack reform process. of women’s mobility, women’s limited access to government institutions responsible for land administration, and lack of The KP government is committed to increasing women’s knowledge, among others (Khan, Khan, and Zahir 2020; Mum- access to services, as manifested by its hiring, for the taz and Noshirwani 2013). Yet land comprises a large share first time, 12 female patwaris (field officers responsible of wealth for rural households, and residential property own- for maintaining land records) and female inspectors and ership accounts for the largest source of growth in wealth in property tax collectors.7 The newly recruited female patwar- Pakistan (Joubert and Kanth 2022). The gender gaps in prop- is are highly educated, some having bachelor’s degrees, and erty ownership could therefore have negative consequences are expected to play a key role in administering land, particu- for women’s wealth accumulation. larly for women.8 All patwaris (male and female) will receive training in Geographic Information Systems (GIS) and infor- The digitization of KP government’s tax administration mation technology in view of the digitization of land records.9 services presents new opportunities and challenges for In addition, the BOR and ETNCD are considering estab- women. Land records in the BOR specify the ownership and lishing within field tax offices female counters—desig- are used for tax purposes, and the urban immovable property nated desks staffed with female tax officials—to provide tax (UIPT) records are often used to record property owner- services to women.10 ship in urban areas (World Bank 2020).6 Digitization of these 4. This is an annual road tax that motor vehicle owners pay. 5. The study also collects data on the agricultural income tax and professional tax. It does not, however, include the general sales tax on services (GSTS) administered by the KP Revenue Authority (KPRA), because there is an ongoing study by KPRMP looking at the gender implications of the GSTS. See Appendix A for the KP provincial tax revenue for the fiscal years 2018–2020. 6. A standard method is lacking for recording of property rights in urban areas, where urban property records may be maintained by different agencies, such as the Excise, Taxation, and Narcotics Control Department (ETNCD) in the UIPT, housing development agencies, and cantonments (World Bank 2022a). 7. See https://revenue.kp.gov.pk/patwar-culture-reforms/. 8. This information was obtained during a key informant interview with DD BOR. 9. See https://revenue.kp.gov.pk/patwar-culture-reforms/. 10. This information was obtained during discussions with senior officials of BOR and ETNCD. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 10 Understanding women’s constraints and experiences in The study is organized is follows: section 2 summarizes the property registration and tax payments and the gender functions of the provincial tax offices and ongoing reforms in gaps in access to and use of digital technologies can im- KP Province; section 3 discusses the survey data collection prove the ability of revenue administrations to address methods; and section 4 considers the conceptual framework the differing needs of women and men.  In KP province, linking the contexts on women’s constraints and experiences this is particularly crucial in the context of ongoing reforms. in property registration and tax payments with tailored mea- Taxpayer programs can assist women in property registra- sures for improving women’s voluntary tax compliance; sec- tion and tax payments, which in turn can help the KP gov- tion 5 shows the results on women’s constraints and experi- ernment increase own-source revenue. Facilitating women’s ences, including at provincial tax offices; section 6 provides a property registration may also lead to other benefits; evidence summary of measures that could facilitate women’s voluntary around the world has shown that increasing women’s prop- tax payments and improve their access to services; and sec- erty rights can help them generate or diversify their incomes, tion 7 draws conclusions on the report’s findings. provide financial security, or serve as collateral for access to finance  (Deere and Doss 2006;  Meinzen-Dick et al., 2011; Pradhan, Meinzen-Dick, and Theis 2019).11 It could also help improve different dimensions of family well-being (Allendorf 2007; Deininger, Goyal, and Nagarajan 2013; Doss 2006; Mishra and Sam 2016; Wang 2014). 11. It is important to consider the contexts and different dimensions of property rights; women’s legal ownership of land may not guarantee that women can exercise their rights to the property, including to bequeath, sell, or use the proceeds from the property (Doss and Meizen-Dick 2020; Pradhan, Meinzen-Dick, and Theis 2019). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 11 2. >>> Provincial Tax Offices and Ongoing Reforms in KP Province The two tax agencies of BOR and ETNCD have a Secretariat in Peshawar and field offices else- where in the province, including service delivery centers (SDCs) and excise and taxation offices with taxpayer facilitation counters. In some districts, ETNCD provides taxpayer services at its separate facilitation centers. 2.1. BOR: Service delivery centers (SDCs) The BOR collects land revenue, the agricultural income tax, and fees on land registration and mutations (a process to make changes to the legal heirs on land title) (Ali and Nasir 2010).12 It is also mandated to administer all land-related matters and to prepare land records, which stipulate ownership and are used for tax purposes. At the subdistrict level (tehsil), tehsildars collect land revenue and other fees (Ali and Nasir 2010). At the local level (Patwar Circle), local revenue officers—patwaris—keep the land records at the lowest revenue areas (Ali and Nasir 2010). Patwaris can prepare FARD (a certificate of land ownership) for a fee and make corrections of errors in the land records (Ali and Nasir 2010). They play a key role in maintaining land transac- tion information and registers of land transfers, but the process of transfers is determined by the district-level deputy commissioners, who are supported by various land administration officers (Ali and Nasir 2010). Significant progress has been made in computerizing land records and land registration, and issuance of land records and transactions can now be processed through the service delivery centers (SDCs) under the purview of the Board of Revenue.13 The BOR runs 28 SDCs across the province, which provide three main services for a fee: (i) issuance of FARD, a certificate of ownership of land containing the names of owners; (ii) mutation, the transfer of ownership through sale, purchase, or inheritance; and (iii) fard-e-badar, the correction of errors on property deeds.14 12. Please refer to BOR’s website for further information about its key functions; https://revenue.kp.gov.pk/. 13. See https://revenue.kp.gov.pk/patwar-culture-reforms/. 14. This information was obtained during a key informant interview with DG ETNCD. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 12 The BOR is reviewing the roles of the patwaris, who so far female patwaris to provide women-dedicated services have must manually maintain land records, by training them in Geo- helped increase the number of landholdings with women’s graphic Information Systems (GIS) and information technol- names on the land titles (World Bank 2022a). The KP govern- ogy in anticipation of the digitization of land records and land ment is attempting to establish a similar initiative to facilitate registration system.15 The computerization of land records is women’s property ownership. expected to promote the transparency of the land registration system and to increase women’s property rights, since women On the legal framework, the KP Enforcement of Women’s will be able to verify their information on land titles more eas- Property Rights Act 2019 was passed to protect women’s ily (Khan, Khan, and Zahir 2020). The recent recruitment of property rights by ensuring that women have access to fast twelve highly educated female patwaris and female tehsildars redress procedures in cases of property disputes, fraud, or through the public service commission will help facilitate the forgery. The property-related appeals and inquiries that the registration of women’s properties and collect revenue.16 In the BOR handles are often related to inheritance, property shares, neighboring Punjab province, the digitization of land records, disputes, or illegal action on the property. establishment of local land registry offices, and recruitment of 2.2. ETNCD: Excise and taxation field offices providing taxpayer facilitation services The ETNCD collects the urban immovable property tax these initiatives took place, only about 27 percent of vehicles (UIPT), motor vehicle registration fee, motor vehicle token tax, owned by KP residents were registered with the province; the professional tax,17 and provincial excises. It provides taxpayer rest were registered with other provinces.20 Motor vehicle own- facilitation services through its field offices in 25 districts of the ers must pay the motor vehicle token tax and registration fee, province, where taxpayers can access services on motor ve- which can be paid at the office or facilitation center of ETNCD hicle registrations and related taxes, the UIPT, and other taxes (henceforth referred as TFCs in this report for ease of refer- under their jurisdiction.   ence) or through the electronic payment system, the Zama KP gateway.21 The KP government has developed the Zama KP Several new initiatives are ongoing to increase the number gateway, a mobile application that allows taxpayers to make of vehicle registrations and facilitate motor vehicle-related tax electronic payments of certain provincial taxes, including the payments in the province. For registration or transfer of own- motor vehicle token tax, on their phones.22  ership of vehicles, the owners or their agents must bring docu- mentation and the vehicles to the TFC for physical verification, but the vehicle records have now been digitized and vehicle For the UIPT, the ETNCD currently sends taxpayers the tax owners will soon be able to register their vehicles online.18 The invoice based on its assessment of the tax due, which can be ETNCD introduced the one-rupee registration scheme and paid through designated banks.23  In future, taxpayers will be universal number plates—promoted in print media and so- able to pay the UIPT through the Zama KP mobile application cial media—to encourage registration of vehicles in KP prov- when the ongoing process of digitization of the urban immov- ince and has piloted a centralized Motor Vehicle Registration able property tax (UIPT) records and linkage to the Manage- System (MVRS) in Peshawar district in 2022, resulting in the ment and Information System (MIS) and GIS is completed. registration of 16,872 vehicles by September 2022.19 Before 15. See https://revenue.kp.gov.pk/patwar-culture-reforms/. 16. See the list of candidates in https://revenue.kp.gov.pk/successful-patwar-candidates/. 17. Professional tax is collected from individuals engaged in health and education, services, finance, trade, business, and government administration earning above Rs.20,000 a month; www.kpexcise.gov.pk/professional-tax. 18. This information was obtained during a key informant interview with DG ETNCD. 19. Based on information shared during the key informant interview with the Director General of ETNCD.  20. See https://tribune.com.pk/story/2307779/one-rupee-car-registration.  21. The registration fee is one-time fee, while the token tax is paid on an annual basis.  22. As per the information shared by KP-ETNCD, until September 2022, Rs.16.65 million has been collected through Zama KP. 23. Based on information shared during the key informant interview with the Director General of ETNCD.  GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 13 3. >>> Data and Methods This study collected data through the following quantitative and qualitative methods: 1. a taxpayer survey of approximately 1,200 respondents, 33 percent of whom are women; 2. focus group discussions (FGDs) with female respondents and female staff at BOR and ETNCD; and 3. key informant interviews (KIIs) with senior provincial tax officials at BOR and ETNCD. The survey questionnaire as well as the FGD and KII instruments were designed by the World Bank team and reviewed by KP government officials and experts. The survey team, Gallup Pakistan, also provided practical suggestions based on their past survey experience in KP. The survey questionnaire was phrased carefully and tested to elicit the roles of women and men in tax payments, their experiences and challenges faced with tax payments and registration, ac- cess to mobile phones, online experiences, and women’s mobility. The survey questionnaire and FGD instruments were translated into Urdu and Pashto. Given that 60 percent of the survey was expected to be administered by telephone, the questionnaire was developed so that the survey would take no more than 15 minutes. To conform to the local norms of limited interactions between nonfamily members of the opposite sex, the enumerators’ sex matched that of the re- spondents, with female enumerators interviewing female respondents and male enumerators in- terviewing male respondents. Gallup Pakistan piloted the survey instrument and conducted the data collection for the taxpayer survey, focus group discussions, and key informant interviews. The project director and staff from the IDA-funded Khyber Pakhtunkhwa Revenue Mobilization and Public Resource Management Program (KPRMP) also participated and assisted in planning and implementing the study.  3.1. Data collection for the taxpayer survey The taxpayer survey used mixed methods to reach different types of male and female respondents through: 1. in-person interviews of women and men visiting field tax offices; 2. telephone interviews of individuals in the provincial taxpayers’ data for the motor vehicle taxes, FARD, and land mutation fees through random sampling; and 3. telephone interviews of female clients of the National Rural Support Program (NRSP). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 14 The mixed methods are useful because taxpayers in the da- and thus are more likely be taxpayers. While implementing tabase and those visiting the field offices might differ in char- the survey, the survey team encountered several challenges acteristics, and identifying respondents through various ap- in interviewing women in the taxpayer database, which are proaches provides a wider pool of taxpayers for the study. In discussed in Box 1. Consequently, it was not possible to inter- addition, following discussions with provincial government of- view women identified from the government’s data, resulting ficials, it was anticipated that not many women would visit the in adjustments to the sampling strategy (provided in Appendix tax offices because of their restricted mobility and prevailing B). The field work for the taxpayer survey started in May 2022 gender roles on tax-related matters. Therefore, it was decided and was completed in July 2022. A workshop was held in Pe- to conduct telephone interviews of taxpayers identified from shawar in October 2022 to present the preliminary findings the provincial government databases and through the NRSP and receive feedback from senior officials from BOR, ETNCD, to increase the diversity of female respondents. The inclusion KPRA, KPRMP, Finance Department, the KP Information and of women in the NRSP is important because they are more Technology Board, and some of the female government of- likely than other women to be in paid employment (such as by ficials who had participated in the FGDs.   engaging in home-based work or managing small enterprises) > > > Box 1 - Challenges faced in implementing the taxpayer survey While implementing the survey, the team encountered the following challenges: 1. The sex of the taxpayer was not indicated in the provincial taxpayer data. The BOR and ETNCD had to manually assign the sex of the respondents based on the taxpayers’ names before respondents could be ran- domly selected. Further, the data on FARD and mutation fee payers was not digitized, making it time-consuming for BOR to collect and compile the data from different field offices. 2. Female respondents in the taxpayer database were difficult to reach by phone. The survey was only able to interview four women from this data source. The reasons more women could not be reached include the following: First, female taxpayers’ phone numbers in the data were usually their male relatives’ numbers, because women were reluctant to provide their own numbers. Second, there was male gatekeeping, as male family members would often not allow women to be interviewed. Third, women were reluctant to answer questions or felt they did not know enough about taxes. Consequently, an additional effort had to be made to increase the participation of women visiting field tax offices or through the NRSP. 3. Female respondents were difficult to interview in person. Due to women’s reluctance or the limited number of women visiting these tax facilities, finding women to interview took longer than anticipated. It was also not pos- sible to interview women at certain offices (namely TFCs in Mardan and Abbottabad) because hardly any women visited there. The BOR and ETNCD assigned focal persons in the field offices (at TFCs or SDCs) with whom the survey firm coordinated interview dates and logistics. These focal persons played a key role in reaching out to the female respondents who were interviewed at these offices. The focal persons also provided office spaces so female or male respondents could be interviewed in private. 4. For all female respondents interviewed in person, no woman came alone to the tax offices; all women were ac- companied by a male family member. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 15 Table 1 shows the number of interviews conducted, by modal- > > > ity. There are 1,153 respondents, of whom 66 percent are TABLE 1 - Number of interviews conducted, by method men and 34 percent are women, which is close to the target Men Women Total of reaching 40 percent women and exceeds the target of 1,000 In-person interviews 1 81 237 318 respondents. Twenty-eight percent were in-person interviews, Tax facilitation center 53 18 and 72 percent conducted by telephone. There is a gender Service delivery center 0 100 difference in the distribution of in-person versus telephone in- Government or tax office 6 106 terviews: the proportion of men interviewed by phone is Respondent’s home or office 22 5 89 percent compared to 40 percent of women, due to the difficulty of reaching women by phone. For the face-to-face Other 0 8 interviews, the agreement with BOR and ETNCD was to cover Telephone interviews 678 157 835 survey sites in (a) locations with heavy traffic, namely Pesha- Taxpayer database for 179 2 motor vehicle taxes2 war and Mardan, given that more female taxpayers were likely to be found there; and (b) one other location. (See Appendix Taxpayer database for 499 2 FARD/mutation C for more details on the survey methodology.) The disag- NRSP clients 0 153 gregated number of respondents for the taxpayer survey and FGDs by districts and modality are given in Appendix D.  Grand total 759 394 1,153 Source: Taxpayer Survey. Note: 1. All in-person respondents for UIPT and FARD/mutation taxpayers in Pe- There is no gender difference in the district of residence distri- shawar were interviewed in government tax offices, challan offices, or re- spondents’ homes. bution, except in Kohat, where the proportion of men is higher, 2. Phone numbers for taxpayers of motor vehicle registration and token tax were available for Peshawar and Abbottabad, but not for Mardan. and in Mansehra, where the proportion of women is higher (see Table 2). Approximately 43 percent of respondents are from Peshawar; 25 percent are from Mardan; and 20 percent > > > of men and 15 percent of women are from Kohat. There is TABLE 2 - District of residence (% of respondents) also no gender difference in the percentage of rural respon- (1) (2) (1)–(2) dents. The literacy rates are higher for respondents in the Men Women Difference survey than the provincial average, and more so for women Peshawar 42.2 43.4 -1.2 (see Appendix E for discussion of individual characteristics). It Mardan 22.3 25.6 -3.3 is possible that individuals in the provincial tax net are better Kohat 20.6 14.5 6.1*** educated than the general population because they must be property owners to be in the tax net for property-related taxes, Abbottabad 3.8 2.5 1.3 and wealthier individuals may be more educated.  Mansehra 1.3 5.6 -4.3*** Other districts 9.9 8.4 1.5 Rural 36.1 35.0 1.1 Number of respondents 759 394 Source: Taxpayer Survey. Note: *** The difference in means is significant at 1%. 3.2. Focus group discussions and key informant interviews Four focus group discussions (FGDs) were conducted ments. For the female staff, additional questions were asked with female respondents (two in Peshawar and two in Mar- about what could be done to improve services to women, what dan). In total, 35 women participated in the taxpayer FGDs constraints female staff faced in carrying out their work, and (see Appendix D). There were also two FGDs with female whether there were physical facilities dedicated to women’s tax administrators, namely staff with BOR and ETNCD. A to- use. Two key informant interviews with Deputy Director tal of 15 female staff participated from BOR in Peshawar and of BOR and Director General of ETNCD were organized to ETNCD in Hayatabad. The qualitative discussions involved review existing mechanisms for women and men to access open-ended questions to identify key issues and constraints tax services, measures that could improve women’s property women face in provincial tax registration, tax payments, and registration and tax payments, and their perceptions of female resolving tax-related problems and to outline areas that would taxpayers. make it easier for women to access services and make pay- GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 16 4. >>> Facilitating Women’s Voluntary Tax Compliance by Understanding Women’s Constraints and Experiences An understanding of the context of the constraints and experiences of women registering property and making tax payments and of the gender gaps in access to and use of digital technologies can help the revenue authorities address the needs of both women and men.  Figure 1 shows the conceptual framework linking how an understanding of these contexts can help revenue authorities develop strategies for improving women’s voluntary tax compli- ance. Gender norms are implicit rules—informed by common beliefs and expectations—that establish which behaviors and status are appropriate for women and men, including regarding women’s asset ownership and mobility (World Bank 2019a; World Bank 2021). Gender roles are shaped by these norms, affecting women’s and men’s roles and activities, such as in registering property and paying taxes, accessing and using digital technologies, and obtaining property- or tax-related services and information. Context-specific norms can differ by location, education levels, and incomes (World Bank 2021). The gender roles and differences in experiences at tax offices could, in turn, affect how women and men view the tax administrations and perceive the quality of government services. Several factors contribute to women’s low property ownership in Pakistan. First, women’s lack of access to or engagement with institutions responsible for maintaining land and other property records is one of the main constraints to women owning property (Khan, Khan, and Zahir 2020). This is because women are largely limited in their mobility, sometimes requiring male chaper- ones or permission to visit government offices (Kamal and Woodbury 2016). Women also often perceive institutions responsible for maintaining property records to be male-dominated, and they may give the power of attorney to male family members to register or transfer property on their behalf (Khan, Khan, and Zahir 2020). Yet women may be unable to verify that the property was, in fact, registered in their names, given the lack of access to these institutions (Kamal and Woodbury 2016). Second, women are often unaware of the land registration systems and proce- dures required to have their names registered on the title (Mumtaz and Noshirwani 2013). Third, a woman may be unable to claim her inheritance rights if she is pressured by her natal family to give up her share of property or when the property is transferred from her father to her brothers without her knowledge (Mumtaz and Noshirwani 2013).    GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 17 Improving provision of public digital services and digitization strictions, norms, and childcare responsibilities (Hanmer, Es- of land records may make it easier for women to register prop- quivel-Korsiak, and Pande 2021);24 (iii) women may not have erty by increasing access to records and reducing the time the literacy, numeracy, or digital skills to use mobile services; required for travel or for interactions with government officials. and (iv) cultural norms or family gatekeeping may consider it But despite these opportunities, new challenges may emerge inappropriate or unsafe for women to use digital technologies as women’s lack of access and use of digital technologies (GSMA 2021; World Bank 2021; World Bank 2022b). Under- could exacerbate the gender gaps in getting government standing these context-specific constraints and experiences services. These differences in access to digital technologies (discussed in section 5) is important to develop tailored strate- could arise because (i) devices and connectivity costs are un- gies (discussed in section 6) that meets the differing needs affordable, particularly in Pakistan where the cost of mobile of women and men. Generating and using sex-disaggregated devices is one of the highest in South Asia, partly due to the taxpayer and service user data would help revenue authorities high tariffs on mobile phones (World Bank 2022b); (ii) it may assess performance and feedback and ensure that programs be harder for women to obtain the identification documents re- and services can be continuously improved. quired for access to digital public platforms due to mobility re- > > > F I G U R E 1 - Conceptual framework to facilitate women’s voluntary tax compliance Source: Authors’ elaboration 24. Hanmer, Esquivel-Korsiak, and Pande (2021) identified many obstacles women face in obtaining identification documents in Nigeria, including their husbands or fathers denying them permission to register, lack of mobility due to childcare responsibilities, inability to take time off from work, poor facilities, and long wait times. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 18 5. >>> Understanding the Context: Women’s Constraints and Experiences 5.1. Women’s mobility and access to and use of digital technologies 1. Women, particularly women with limited education, are largely confined to their homes and require permission or a male relative as chaperone to visit government or tax offices. The survey results show that most women (87 percent) need permission from their families to visit government or tax offices (Figure 2). The mobility of women with less education—with primary schooling or less—is more restricted, at 95 percent. Even if accompanied by male relatives, 44 percent and 18 percent of women cannot visit tax of- fices and government offices, respectively (Figure 3), and the mobility of women with limited education is even more restricted. > > > F I G U R E 2 - Need permission from male relatives to visit government or tax office Source: Authors’ calculations based on Taxpayer Survey. Note: Women who are less educated have completed primary schooling or less, and more educated women have completed more than that level of schooling. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 19 > > > F I G U R E 3 - Cannot visit government or tax office even if accompanied by male relatives Source: Authors’ calculations based on Taxpayer Survey. 2. Norms that restrict women’s movements are likely 3. Access to and use of digital technologies vary by to limit their ability to access services and register education; women with limited education are less property or pay taxes. Women in the focus group dis- likely than educated women to own mobile phones cussions highlighted the problems of restricted mobility: and to use digital banking and digital wallet applica- they were not allowed to go to government offices be- tions. Only 35 percent of less educated women (primary cause the offices were male-dominated spaces. As a re- schooling or less) own mobile phones, compared to 66 sult, they were not able to register property or vehicles percent of all women (Figure 4). Similarly, a smaller share themselves. Women needed male relatives to register of women with limited education have used mobile bank- property on their behalf. In other cases, women did not ing apps (9 percent) and mobile wallets (17 percent) than register property because they were not permitted to go to women with more education (23 percent and 63 percent, the government office. Some women also indicated they respectively) (Figure 5). Connectivity to the internet is needed men to accompany them to be taken seriously at also an issue for less educated women and men, where government offices. 46 percent of men and 38 percent of women with primary schooling or less do not have internet access (Figure 5). In Pakistan, the internet is accessed mostly through mo- Excerpts from focus group discussions bile devices, and the unaffordability of the devices and of “Transfer of property in my name was difficult. Men of my connectivity, as well as connectivity challenges in smaller family said you would not go (to the government office) so it is still registered in my uncle’s name.” [Female tax- cities and rural areas, may contribute to these results payer] (World Bank 2022b). Broadband penetration is low, with “The men in our house do not allow us to go (to the gov- fixed broadband investments being concentrated in larger ernment office) so the registration was done by my hus- cities (World Bank 2022b). band.” [Female taxpayer] “My brother was the one who did everything. I did not go 4. While there is no gender gap in mobile ownership, inside. Only went for the signature.” [Female taxpayer] more men than women are likely to have used mo- “Property is not registered in the name of a female be- bile banking applications (26 percent of men versus cause men say that there will be many problems if they 17 percent of women), highlighting the differences in register in women’s name. Men will be able to visit the digital financial inclusion. Yet women are more likely to offices, so they prefer registering it either by the name of their son or husband.” [ETNCD female tax official] have bought goods online, suggesting that women have used cash on delivery for online purchases. There is also GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 20 some evidence of family gatekeeping for women, with tion services, women with limited education may not 10 percent of women citing family disapproval as the fully benefit from digitization. There are also gender main reason for not purchasing online, compared to 4 per- gaps in digital banking that could limit women’s ac- cent of men (Appendix F). cess to digital platforms for paying taxes. Face-to-face service provision therefore continues to be important, par- 5. These results suggest that while digitization of tax ticularly for less educated women, at least in the medium services—such as electronic registration and tax term, until more women and men have better access to payments on mobile phones—could reduce compli- and use of digital technologies. ance costs and improve access to tax and registra- > > > F I G U R E 4 - Ownership of mobile phones and use of mobile apps, by sex Source: Authors’ calculations based on Taxpayer Survey. Note: The difference in means is significant at *** <0.01, and * <0.1. > > > F I G U R E 5 - Ownership of mobile phones and use of mobile apps, by sex and education Source: Authors’ calculations based on Taxpayer Survey. Note: Less educated individuals attained an education of primary schooling or less. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 21 5.2. Gender roles and methods of tax payment and property registration 1. Men are more likely than women to be responsible for the difference between men and women paying the taxes paying UIPT, motor vehicle taxes (MVT), FARD/muta- is not significant (Figure 6). When we estimate the condi- tion fees, and professional tax. Men also tend to pay tional probability of tax payment, controlling for individual the tax on others’ behalf. The unconditional means for and household characteristics, women are less likely to tax payment show that men are significantly more likely pay all taxes (see Appendix H).25 than women to pay all taxes, except for UIPT, where wom- en are more likely to pay taxes (Figure 6). For the MVT, > > > F I G U R E 6 - Payment of provincial taxes Source: Authors’ calculations based on Taxpayer Survey. Note: The figure shows the percentage of respondents who paid taxes in the prior two years (for all respondents). A person who paid any tax is defined as the respondent who paid UIPT, motor vehicle tax, professional tax, FARD/mutation fee, or agricultural income tax in the prior two years. *** The different in means is significant at 1%. 25. Women interviewed in person in Peshawar were identified by the ETNCD, so they are more likely to pay the UIPT (see Appendix D). The number of women paying UIPT is likely overrepresented in the sample for this reason, which is why a regression analysis is conducted. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 22 2. Taxpayers continue to travel to payment points to pay percent) have used services at SDCs than did their male taxes in person rather than using electronic payment counterparts (43 percent).26 Use of bank applications methods, and men are more likely than women to pay and the Zama KP app to pay taxes is not yet common, taxes at a bank branch. Among taxpayers (respondents much less so for women: 12 percent of men use bank- responsible for paying provincial taxes), payment of taxes ing apps for tax payment versus only 4 percent of women, through banks (challan form) is common for men (51 per- and only 3 percent of men and 1 percent of women have cent) but not as much for women (23 percent) (Figure 7). used the Zama KP app. The use of tax facilitation agents Approximately, half the male and female taxpayers have (such as accountants) is also uncommon. used services at TFCs, while more female taxpayers (60 > > > F I G U R E 7 - Methods used for provincial tax payment or property registration Source: Authors’ calculations based on Taxpayer Survey. Note: The sample is restricted to taxpayers, who are defined as respondents who are personally responsible for paying UIPT, motor vehicle taxes, FARD or mutation fees, professional tax, or agricultural tax. The figure reports the percentage of taxpaying respondents who have used the methods included to pay tax or register property. *** significant at 1%, ** 5%, * 10%. 26. When individual and household characteristics are taken into account in a linear probability model, the gender difference in the use of TFCs or SDCs is no longer signif- icant (see columns 1 and 2 in Appendix G). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 23 5.3. Gender gaps in access to tax information There are gender gaps in how women and men access 2. Social media is an important source of information tax information, highlighting the need to tailor tax cam- for 33 percent of men, compared to only 16 percent of paigns or information strategies to reach different types women, and it is more important for highly educated of taxpayers. than for less educated men and for younger rather than older individuals.27 The conditional probability of 1. Women tend to rely on personal networks—friends using social media for tax information (from a linear prob- and family—to obtain information on taxes, particu- ability model) is 41 percent for men with a bachelor’s de- larly for less educated women. Majority of women (60 gree and less than 30 percent for men without (panel 2 in percent) receive information from friends or family versus Figure 9). Younger individuals are also more likely to use 48 percent of men (Figure 8). This is corroborated in the social media (see Appendix I). FGDs, where women confirmed obtaining tax information from their male family members, such as brothers and 3. There is no gender difference in whether TFCs or husbands. The estimated probability of relying on friends SDCs are chosen as an information source. Using sim- and family for tax information (from a linear probability ple averages in our sample, women are more likely to cite model) reveals that this source is more important for less TFCs or SDCs as important sources of information than educated than more educated women (panel 1 in Figure are men, but when individual and household characteris- 9): the conditional probability that women who rely on per- tics are controlled, there is no longer a gender difference sonal networks did not attend school is 76 percent com- (panel 3 in Figure 9). Education also has no influence on pared to 49 percent for women with a bachelor’s degree. whether the TFCs or SDCs are cited as one of the two Word-of-mouth informal information dissemination from important information sources. their families or people they know plays an important role for less educated women. > > > F I G U R E 8 - Source of information on taxes Source: Authors’ calculations based on Taxpayer Survey. Note: Respondents chose up to two main sources tax information. The gender difference is significant at *** 1%, **5%, *10%. 27. Social media includes use of social networking platforms. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 24 4. Provincial tax authorities’ websites are not a major rently is not available in Urdu or Pashto—it is only source of tax information. Only 11 percent of men and available in English. Translation of materials on the web- 3 percent of women cite the provincial ETNCD and BOR sites into Urdu and Pashto would help with dissemination websites as being their main sources of information. No- of tax information. tably, most information on the provincial websites cur- > > > F I G U R E 9 - Source of information on taxes, by sex and education Source: Authors’ calculations based on Taxpayer Survey. Note: The graphs show the estimated probability of the important source of tax information from a linear probability model by education. The vertical lines show the 95% con- fidence interval. The regression also includes (but not shown here) age, educational attainment, whether respondent can read and write in English, employment status, households’ source of drinking water, mobile ownership, presence of children in household, households with no male adults, rural area, district of residence, whether interview was conducted in Pashto or Hindko, and whether the respondent was interviewed in person. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 25 5.4. Satisfaction with services and challenges at provincial tax offices Level of satisfaction 2. Lack of information and unclear processes at the SDCs are the most-cited challenges for women, as Among users of services at provincial tax offices, women highlighted by 58 percent of female users compared are generally more satisfied with the services at TFCs or to 45 percent of male users (Figure 13). Several FGD SDCs than men (Figure 10). Men tend to express their dissat- female participants, most of whom are highly educated, isfaction with the services: 20 percent of men are somewhat or confirmed problems understanding the forms and proce- completely dissatisfied with services at the TFC, and more so dures due to the complexity of the language used. at the SDC, at 30 percent. Challenges at provincial tax offices Excerpts from focus group discussions on property registration Nevertheless, almost half the male and female users ex- “The language being used during the (property regis- perienced difficulties at the SDCs, and to a lesser extent tration) process is extremely difficult. Either easy Urdu at the TFCs (40 percent of men and 25 percent of women) or Pashto should be used.” [FGD participant] (Figure 10). Women tend to experience longer processes for tax payments or registration than men at both offices 3. Long lines and lack of staff access are also challenging (panels 1 and 2 in Figure 12). The satisfaction levels are cor- for women (28 percent) and men (36 percent) at both of- related with the time spent on processes for men: panels 3 fices (in Figure 13). FGD participants corroborated these and 4 in Figure 12 indicate that most male users who were challenges, citing as key barriers to accessing tax servic- satisfied with the services at TFCs or SDCs spent less than es crowds, long queues, lack of counters to address the two hours compared to 86 percent or more of unsatisfied male processes, and inadequate education of the clerical staff. users who spent more than two hours. 4. Lack of female staff is the second most often cited The following challenges with services at provincial tax offices difficulty affecting female taxpayers’ experience (Fig- were raised by survey respondents and FDG participants. ure 13). A third of female users raised the lack of female staff at offices as a key obstacle at the SDCs, which was 1. The long process for property registration is a chal- also confirmed in the FGDs. lenge. A senior BOR official revealed that in FARD and mutation cases at SDCs, clients had to visit three or Excerpts from focus group discussions on lack four counters for attestation, signatures, and regis- of female staff tration, with each counter performing a separate task. A “Make an office specially for women, with only women token is needed from the queue management system for staff so communication can be done easily. Women each task, which along with the need for documentation to cannot visit office and are unable to convey their mes- verify property rights, makes registration a long and time- sage to the male staff, so then how will they resolve their issues?” [FGD participant] consuming process. Measures to improve services to women based on these chal- lenges are discussed in section 6. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 26 > > > F I G U R E 1 0 - Satisfaction with services among users Source: Authors’ calculations based on Taxpayer Survey. > > > F I G U R E 1 1 - Faced challenges at the facility among users Source: Authors’ calculations based on Taxpayer Survey. Note: *** The gender difference is significant at 1%. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 27 > > > F I G U R E 1 2 - Length of time spent at TFCs or SDCs to pay, register, or solve problems Source: Authors’ calculations based on Taxpayer Survey. Note: The number of female users who were unsatisfied were less then 50 observations, so it was not possible to disaggregate by satisfied or not satisfied. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 28 > > > F I G U R E 1 3 - Types of difficulties experienced at tax offices (as a percentage of users who faced challenges) Source: Authors’ calculations based on Taxpayer Survey. Note: Respondents chose up to two main challenges faced. Female users who faced challenges at TFCs are not shown because there were less than 50 observations. The gender difference is significant at *** 1%, **5%, *10%. 5.5. Gender differences in trust in KP tax administration and reciprocity Trust in KP tax administration We observe two patterns in Figure 14. First, taxpayers have more trust than nontaxpayers: 71 percent of female taxpay- Building trust in tax administrations is an important element of ers have trust in KP tax administration relative to 52 percent fostering voluntary compliance (Dom et al. 2022). The survey of female nontaxpayers. Second, more educated individuals elicits respondents’ views on whether they trust the provincial have trust compared to less educated individuals: 65 percent tax administration, and responses are scaled into five groups: of women with more than primary schooling have trust in the strongly agree, agree, neutral, disagree, strongly disagree. tax agencies compared to only 51 percent of women with less We categorize the indicator as having trust (=1) if respondents schooling. A similar pattern is observed for men. strongly agree or agree that they can trust the KP tax agen- cies. There is no gender difference in the level of trust toward the KP tax administration, a result that holds even when disaggregated by taxpayer status or by education (panels 1 and 2 in Figure 14). The gender difference is not significant overall, among taxpayers or nontaxpayers, and among edu- cated or less educated individuals. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 29 > > > F I G U R E 1 4 - Trust in KP administration Source: Authors’ calculations based on Taxpayer Survey. Note: Less educated are individuals with primary schooling or less. Reciprocity: Tax revenues and government service Women are less likely than men to believe that the gov- delivery ernment is using tax revenues to provide services, and this result holds even when they are classified by tax- An emerging body of literature highlights government service payer status or by education (panels 1 and 2 in Figure 15): delivery as an important avenue for increasing voluntary tax 61 percent of men perceive reciprocity, compared to only 48 compliance in low- and middle-income countries (Carrillo, percent of women. If women do not believe the government is Castro, and Scartascini 2021; Gonzalez-Navarro and Quin- using taxes to deliver services, this may lead to lower tax mo- tana-Domeque 2016; Kresch et al. 2023). We use the term, rale and lower tax compliance—women may be less inclined reciprocity—defined in Dom et al. (2022)—to measure the to register their property or businesses with the government extent to which respondents believe that the government is and decide to stay informal. using tax revenues to provide services to its citizens. The re- sponses are scaled into five groups: strongly agree, agree, neutral, disagree, strongly disagree. We define respondents to have perceived reciprocity (=1) if he/she strongly agrees or agrees with the statement. > > > F I G U R E 1 5 - Perceived reciprocity: Agrees that taxes are used for improving service delivery Source: Authors’ calculations based on Taxpayer Survey. Note: Less educated are individuals with primary schooling or less. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 30 6. >>> Measures to improve women’s tax payments and registration This section discusses efforts or measures that could help improve female taxpayers’ experi- ences in registration and payments, drawing from the analysis of the taxpayer survey, insights received from the FGDs and KIIs, and review of the literature. Appendix J provides a summary of the recommended actions that are consistent with the KPMRP action plan and results frame- work. 6.1. Availability of female tax officials and establishment of dedicated female counters Female taxpayers believe more staff and availability of female tax officials would make it easier to pay taxes and register property. Forty percent of female taxpayers would like more staff, and 23 percent would like female staff at TFCs or SDCs (Panel 1 in Figure 16). There are indeed very few female staff stationed at the field tax offices. In response, both senior officials from BOR and ETNCD suggested establishing counters with female tax officials and wait- ing rooms dedicated to serving female clients in larger offices. Once established, an infor- mation dissemination campaign should be conducted about the availability of female counters and the types of services provided to citizens. Such efforts are likely to promote women’s property rights and facilitate voluntary taxpayer com- pliance. In the neighboring Punjab province, many female patwaris run village-level offices for land revenue and emergency services, known as Dehi Markaz-e-Maals, and the provincial Pun- jab government aims to establish such offices in every tehsil (World Bank 2022a). There are also female revenue officers in many Arazi Records Centers (local land record offices) who service female-only counters (World Bank 2022a). In rural Punjab province, computerizing land records, increasing access to the records, and establishing female counters at village and local-level land offices have helped facilitate registration in women’s names of 30 percent of land in areas where gender measures were in place (World Bank 2022a). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 31 6.2. Safe and gender-equitable work environment 1. Establishing counters at field offices dedicated to and male staff. Additional training may be needed for fe- serving women and, in the case of the BOR, shifting male staff to give them the skills needed to play a greater functions to SDCs require reviewing the institutional role in providing tailored property registration and pay- structure and female tax officials’ roles and authori- ment services to female taxpayers. ties, including female patwaris and tehsildars. Further, computerizing land records and digitizing tax registration 3. A safe and equitable workplace environment—includ- and services provided by BOR and ETNCD may mean ing anti-harassment, flexible working arrangements, that roles of male and female personnel will need to be re- childcare policies, and physical safety—is important viewed. According to the FGDs with staff, female officials’ to attract and retain female staff (Schomer and Ham- work was often restricted to basic administrative tasks, mond 2020). It is recommended to review safety policies and women were given limited authority, requiring them to and protocols, childcare policies, and working practices confer frequently with their managers for direction. and to develop anti-harassment guidelines to make sure that women’s needs are met (Schomer and Hammond 2. It is recommended that gender equitable promotions 2020). The following concerns were raised in the FGDs and training be considered an essential component with women: (1) the Anti-Harassment Bill was passed, of improving services to female property owners and but there were no clear guidelines on implementation; (2) taxpayers. Female officials at BOR and ENTCD com- there were safety concerns over women’s line of work; plained about not being given the same opportunities and (3) there were instances in which some meetings for training and promotion as male staff. To move toward did not end at regular working hours, making it difficult to computerized land records and public digital services re- maintain work-life balance. quires training on digital technology use for both female 6.3. Better facilities—washrooms and separate areas for women To retain female staff and improve services to female Excerpts from focus group discussions on lack of taxpayers, it is important to establish separate bathroom women’s bathrooms facilities for women. New infrastructure investments planned for establishing the one-stop TFCs should in- “If we ever want to go to the washroom, we go to McDon- alds or Pizza Hut.” [ETNCD female staff] clude separate washroom and other facilities (such as space for eating and praying and lactation rooms) for “There was only one washroom in a meeting hall. And when we used to go there, the guard used to stop and female staff and clients. The lack of facilities for women is say that the meeting is going on and you cannot go in- a main concern for female tax officials, where one official re- side. Then where should I go? I used to go to McDonalds.” vealed that because there were no washrooms for women in [ETNCD female staff] the office, she had to use the facilities at a nearby fast-food “We do not have any space to pray, and we don’t even restaurant. have a washroom.” [ETNCD female staff] Better facilities are also a key concern for male and female taxpayers (36 percent of men and 28 percent of women in panel 1 of Figure 16). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 32 6.4. Simplification of processes and clearer guidance It is important to review the implications of the registra- single female employee can carry out all three tasks of land tion and payment processes of reforms involving digi- registration (namely attestation, signatures, and registration) tized services, computerizing databases, and service at the same desk so that the clients do not have to go through counters for women at SDCs and tax facilitation centers. different counters and join the queue each time. Computerization of land records and digitization of vehicle registration and UIPT payment are likely to shorten and sim- Further, female FGD participants, many of whom are highly plify the process for property registration and payment, as was educated women, recommended that easy Urdu or Pashto the case in Punjab province.28 If service counters dedicated to be used for guidance on tax payments and registration women are established, a senior BOR official proposed that a forms to address the problem of unclear rules and processes. 6.5. Tax officials who speak the local language Having staff who speak the local language is important, conditional probability of staff speaking the local language as particularly for less educated women: 35 percent of female important is higher for less educated than for more educated nontaxpayers want staff who speak the language they under- women: 36 percent for women with no education and less stand versus 18 percent of men (panel 2 of Figure 16). The than 18 percent for women with more education (Figure 17). 6.6. Use of digitized services—electronic registration and tax payments 1. The gender differences in (i) preferences for elec- an ambitious electronic governance initiative that includes tronic tax payments; and (ii) access and use of digital provision for digital public services.29 The federal govern- technologies and digital financial services suggest ment is also embarking on a program to invest in digital that women, particularly with less education, will con- public infrastructure and services (World Bank 2022b). tinue to need in-person interactions and support for Parallel efforts to increase women’s digital literacy and fi- tax payments, at least in the medium term. The results nancial capacity are also needed to encourage women’s show that men are more likely than women to be willing use of digital public services (World Bank 2022b). to pay taxes electronically: A quarter of male taxpayers would like to be able to make electronic tax payments ver- 3. Successful implementation of electronic tax payment sus only 9 percent of female taxpayers (panel 1 of Figure efforts requires consistent taxpayer education and 16). The predicted probability of preferring to pay online support and regular communication and feedback (in panel 2 of Figure 17) shows that men with at least pri- from taxpayers (Dom et al. 2022; Okunogbe and Santoro mary education are more likely than women to prefer this 2022). While electronic registration, filing, and payments method of payment. can reduce tax compliance costs because taxpayers do not have to travel to payment centers or have in-person 2. A differentiated strategy for digital access and com- interactions with tax officials, taxpayers need to be con- munication for women is important to enhance their sistently educated, facilitated, and supported to use the use of digital services. The strategy should include applications (Dom et al. 2021; Okunogbe and Santoro making digital devices and connectivity more affordable, 2022). The Zama KP app is available in three languages simplifying applications and content, and designing con- (English, Urdu, and Pashto), but it is only available for tent and services tailored to women (World Bank 2022a; Android users, not for iOS users. While the majority of us- World Bank 2021). The government of KP is embarking on ers in Pakistan are Android users and only 21 percent are 28. In Punjab province, the Land Record Management Information System, with the support of the World Bank, shortened the process of land registration and mutations (Khan, Khan, and Zahir 2020). 29. This initiative is being supported under KPRMP. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 33 estimated to use iPhones,30 it is important to be inclusive (panel 1 of Figure 16). A senior government official from of all users. The reviews of the app also suggest that it BOR stated that a proposal for a call center and a citizen needs more consistent updating and improvement. facilitation center at the divisional level were in the works through the Khyber Pakhtunkhwa Information Technology 4. Call centers and online help desks would be preferred by Board (KPITB). 17 percent of male and 18 percent of female taxpayers 6.7. Tax education and information campaigns on women’s property rights It is recommended that strategies on tax education and in the FGDs). Sensitization campaigns can target both women awareness campaigns be tailored to the different needs and men, who could play a role in assisting women’s property of taxpayers. Different social media channels can be used registration. Relatedly, it is also important to train male rev- to disseminate tax-related information for more educated and enue officers on gender issues in property rights to reduce younger women and men. In contrast, information disseminat- gender biases in property registration (World Bank 2022a). ed through field-level efforts—such as TV or radio programs or provided by field staff in local languages—are important for in- Presently, provincial excise department and BOR websites, dividuals with limited education. Partnerships with community- which do not serve as the main source of tax information, are based organizations could be established to organize tax edu- only available in English. Translation of tax information on cation or awareness campaigns for women on the importance these websites into Urdu or Pashto could provide informa- of property rights, land registration processes, and taxpayer tion to a broader audience. services, including where to get these services (as suggested 6.8. Compilation, digitization, and sex-disaggregation of taxpayer and service user data Digitizing and compiling taxpayer data among the three Generating and using sex-disaggregated taxpayer and provincial tax administrations and sex-disaggregation of service user data could help tax administrations develop data would allow the tax agencies to track progress in tailored communication and taxpayer programs to meet property registration and tax compliance. Currently, the the needs of both women and men. It could help the gov- revenue and visitor data on mutation and FARD fee payers ernment assess performance and feedback and ensure that and tax service users are not digitized and are managed sepa- programs and services can be continuously improved. It could rately in each field office. The visitor data is also not sex dis- also help the KP government to determine which one-stop aggregated. The lack of digitization and centralization of tax- TFC could establish the service counters dedicated to payer registries limits information sharing and the use of this servicing women and help in directing human and finan- information for monitoring tax compliance and planning. There cial resources for this purpose. Along with digitization of are also duplicate entries of taxpayers in the database, and data, it is recommended that training be provided to tax of- the database for some taxes (such as the urban immovable ficials on how to use the technology and manage the data. property tax and FARD/mutation) do not identify the sex of the taxpayers. The business process reengineering and integra- tion of the taxpayer databases are important output indicators in the IDA-funded Khyber Pakhtunkhwa Revenue Mobilization and Public Resource Management Program (KPRMP) that would support improvement in tax compliance and efficiency in tax administration (World Bank 2019b). 30. See https://pas.org.pk/smart-phone-usage-in-pakistan-infographics/. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 34 6.9. Physical access To improve access to services, one-stop TFCs should (panel 2 of Figure 16). Physical access is also important for fe- be established in all districts so that taxpayers need not male taxpayers (13 percent versus 8 percent of male taxpay- travel to districts outside of their resident district. Better ers) (panel 1 of Figure 16), as tax payment centers are often physical access to the offices is the biggest priority in mak- far from their homes. A woman in Mardan shared that she has ing it easier to pay taxes for male nontaxpayers (40 percent) to travel to Peshawar to pay taxes. > > > F I G U R E 1 6 - Measures that would make it easier to pay tax or register property Source: Authors’ calculations based on Taxpayer Survey. Note: Respondents chose up to two main measures. The gender difference is significant at *** 1%, **5%, *10%. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 35 > > > F I G U R E 1 7 - Preference for staff who speak the language they understand (panel 1) or for paying online (panel 2), by sex and education Source: Authors’ calculations based on Taxpayer Survey. Note: The graph presents the estimated probability of respondents wanting staff who speak the language they understand (panel 1) and for making payments online (panel 2), from a linear probability model for all respondents. The vertical lines show the 95% confidence interval. The regression also includes (but not shown here) age, educa- tional attainment, whether respondent can read and write in English, employment status, households’ source of drinking water, mobile ownership, presence of children in household, households with no male adults, rural area, district of residence, whether interview was conducted in Pashto or Hindko, and whether the respondent was interviewed in person. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 36 7. >>> Conclusion The KP provincial government is undertaking reforms to facilitate voluntary taxpayer compliance by establishing one-stop taxpayer facilitation centers (TFCs) to assist taxpayers with registra- tion and payment and to coordinate among different provincial tax agencies under one roof. It is also digitizing land records and taxpayer services and establishing common digital platforms for taxpayers to register, file, and make payments online. This study’s objective was to support the KP provincial government’s efforts to facilitate women’s voluntary tax compliance by understanding women’s constraints and their experiences in tax payment and property registration and the gender gaps in access to and use of digital technolo- gies in the context of the reforms. By conducting a taxpayer survey of 1,200 respondents (a third of whom were women), focus group discussions with women and female tax officials, and key informant interviews, the study identified specific challenges women faced in property registra- tion and tax payments. It also revealed that women’s challenges were not monolithic, but varied by education; therefore, tailored measures will be important to meet the needs of different types of taxpayers. While digitization of tax services could reduce tax compliance costs for taxpayers, women with limited education may not fully benefit from digitization because they tend to lack ownership of mobile devices and to have limited use of digital banking and wallet applications. They are also more restricted in their mobility and rely more heavily on personal networks for tax-related infor- mation. For this reason, dissemination strategies on property registration and taxpayer informa- tion would need to reach different types of taxpayers: social media could be used for educated women and men, but a more ground-level tax campaign—using NGOs, TV, or radio—may be more effective for women with less education. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 37 The importance of having female staff provide tax and property ment (e.g., anti-harassment policies and promotion, flexible registration services dedicated to women, simplifying property work, safety protocols), and physical infrastructure—including registration processes, and using easy Urdu and Pashto were bathrooms and separate areas for women. highlighted by female taxpayers. Establishing dedicated ser- vices for women by assigning female tax officials in one-stop Finally, digitizing and sex-disaggregating taxpayer and service TFCs could also help women register property and pay taxes. user data could help revenue authorities develop tailored pro- But to retain staff, these changes require allocation of human grams and services that meet the needs of women and men and financial resources, review of staff roles, additional train- and monitor progress. Such data would be useful in guiding ing, instituting a gender-equitable and safe workplace environ tax administrations to allocate financial and human resources. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 38 References Ali, Z., and A. Nasir. 2010. “Land Administration System in Pakistan—Current Situation and Stakeholders’ Perception.” Pa- per presented at XXIV FIG Congress 2010, “Facing the Challenges—Building the Capacity.” Sydney, Australia. https:// www.fig.net/resources/proceedings/fig_proceedings/fig2010/papers/fs03f/fs03f_ali_nasir_3901.pdf. Allendorf, K. 2007. “Do Women’s Land Rights Promote Empowerment and Child Health in Nepal?” World Development 35 (11): 1975–88. Carrillo, P. E., E. Castro, and C. Scartascini. 2021. “Public Good Provision and Property Tax Compliance: Evidence from a Natural Experiment.” Journal of Public Economics 198: 104422. Deere, C. D., and C. R. Doss. 2006. “The Gender Asset Gap: What Do We Know and Why Does It Matter?” Feminist Economics 12 (1-2): 1–50. Deininger, K., A. Goyal, and H. Nagarajan. 2013. “Women’s Inheritance Rights and Intergenerational Transmission of Resources in India.” Journal of Human Resources 48 (1): 114–41. Dom, R., A. Custers, S. R. Davenport, and W. Prichard. 2022. Innovations in Tax Compliance: Building Trust, Navigating Politics, and Tailoring Reform. Washington, DC: World Bank. License: CC BY 3.0 IGO; https://openknowledge.world- bank.org/handle/10986/36946. Doss, C. 2006. “The Effects of Intrahousehold Property Ownership on Expenditure Patterns in Ghana.” Journal of African Economies 15 (1): 149–80. Doss, C., and R. Meinzen-Dick. 2020. “Land Tenure Security for Women: A Conceptual Framework.” Land Use Policy 99: 105080. Global Partnership for Financial Inclusion (GPFI). 2020. Advancing Women’s Digital Financial Inclusion. Bruxelles, MB, Canada: Global Partnership for Financial Inclusion. https://www.gpfi.org/sites/gpfi/files/sites/default/files/saudig20_ women.pdf. Gonzalez-Navarro, M., and C. Quintana-Domeque. 2016. “Paving Streets for the Poor: Experimental Analysis of Infra- structure Effects.” Review of Economics and Statistics 98 (2): 254–67. GSMA (Groupe Speciale Mobile Association). 2021. Addressing the Mobile Gender Gap in Pakistan. London, UK: GSMA. https://www.gsma.com/mobilefordevelopment/resources/addressing-the-mobile-gender-gap-in-pakistan/. Hanmer, L., V. Esquivel-Korsiak, and R. Pande. 2021. Barriers to the Inclusion of Women and Marginalized Groups in Nigeria’s ID System: Findings and Solutions from an In-Depth Qualitative Study. Washington, DC: World Bank. Joubert, C., and P. Kanth. 2022. “Life Cycle Savings in a High-Informality Setting.” Policy Research Working Paper 10121, World Bank, Washington, DC. Kamal, A., and L. Woodbury. 2016. “Emerging Opportunities for Women in Khyber Pakhtunkhwa: Industry (Manufacturing and Construction), Mining, Tourism, and Agriculture.” Working paper, International Growth Center, Pakistan Program, Lahore University of Management Sciences, Lahore. https://www.theigc.org/publications/emerging-opportunities-wom- en-khyber-pakhtunkhwa-industry-manufacturing-and. Khan, H. H., A. Khan, and H. Zahir. 2020. Women’s Inheritance Rights: Procedures, Loopholes and Obstacles. Lahore, Pakistan: Center for Economic Research in Pakistan. https://www.cerp.org.pk/updata/files/files/46_20200521210833. pdf. Khyber Pakhtunkhwa, Government of, Finance Department. 2023. Actual Receipts and Expenditure Report FY 2020-21. https://www.finance.gkp.pk/article/actual-receipts-expenditure-report-fy-2020-21. Kresch, E. P., M. Walker, M. C. Best, F. Gerard, and J. Naritomi. 2023. “Sanitation and Property Tax Compliance: Analyz- ing the Social Contract in Brazil.” Journal of Development Economics 160: 102954. Mishra, K., and A. G. Sam. 2016. “Does Women’s Land Ownership Promote Their Empowerment? Empirical Evidence from Nepal.” World Development 78: 360–71. Meinzen-Dick, R. S., N.L. Johnson, A.R. Quisumbing, J. Njuki, J. Behrman, D. Rubin, A. Peterman & E.M. Waithanji 2011. “Gender, assets, and agricultural development programs: A conceptual framework.” CAPRi Working Paper. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 39 Mumtaz, K., and M. M. Noshirwani. 2013. Women’s Inheritance and Land Rights in Pakistan. Lahore, Pakistan: Shirkat Gah, Women’s Resource Center. https://shirkatgah.org/shirkat/wp-content/uploads/2017/01/Land-Rights-2nd-Special- Bulletin-2013-English.pdf. NIPS (National Institute of Population Studies) [Pakistan] and ICF. 2019. Pakistan Demographic and Health Survey 2017–18. Islamabad, Pakistan: NIPS; Rockville, Maryland, USA: ICF. Okunogbe, O., and F. Santoro. 2022. “The Promise and Limitations of Information Technology for Tax Mobilisation.” ICTD Working Paper 135, Brighton, England, Institute of Development Studies. DOI: 10.19088/ICTD.2022.001. Pakistan Bureau of Statistics. 2021. Key Findings Report Pakistan Social and Living Standards Measurement Survey: District Level 2019/2020. Islamabad, Pakistan: Pakistan Bureau of Statistics. https://www.pbs.gov.pk/publication/ key-finding-report-pslm-district-level-survey-2019-20. Pradhan, R., R. Meinzen-Dick, and S. Theis. 2019. “Property Rights, Intersectionality, and Women’s Empowerment in Nepal.” Journal of Rural Studies 70: 26–35. Schomer, I., and A. Hammond. 2020. “Stepping Up Women’s STEM Careers in Infrastructure: An Overview of Promising Approaches.” ESMAP Paper, World Bank, Washington, DC. Stantcheva, S. 2022. “How to Run Surveys: A Guide to Creating Your Own Identifying Variation and Revealing the Invisible.” NBER Working Paper Series 30527, National Bureau of Economic Research, Cambridge, MA. http://www.nber.org/papers/ w30527. Wang, S. Y. 2014. “Property Rights and Intra-Household Bargaining.” Journal of Development Economics 107: 192–201. World Bank. 2019a. “Ethiopia Gender Diagnostic Report: Priorities for Promoting Equity.” World Bank, Washington, DC. https:// openknowledge.worldbank.org/entities/publication/91fff14f-197c-50d5-893b-84a86ecc2781. World Bank. 2019b. “Project Appraisal Document on a Proposed Credit for a Program of Results Khyber Pakhtunkhwa Revenue Mobilization and Public Resource Management.” World Bank, Washington, DC. https://projects.worldbank.org/en/projects-oper- ations/project-detail/P162302. World Bank. 2020. “Islamic Republic of Pakistan: Leveling the Playing Field.” Systematic Country Diagnostic, World Bank, Washing- ton, DC. https://openknowledge.worldbank.org/server/api/core/bitstreams/93b61534-2d3b-54c6-89ac-cba6aacb80f3/content. World Bank. 2021. Accelerating Gender Equality Through Digital Technologies. Washington, DC: World Bank. https://thedocs. worldbank.org/en/doc/61714f214ed04bcd6e9623ad0e215897-0400012021/related/Digital-Development-Note-on-Gen- der-Equality-November2021-final.pdf. World Bank. 2022a. “Project Appraisal Document on a Proposed Credit for a Punjab Urban Land Systems Enhancement Project.” Report No. PAD3906. https://projects.worldbank.org/en/projects-operations/document-detail/P172945?type=projects. World Bank. 2022b. South Asia’s Digital Opportunity: Accelerating Growth, Transforming Lives. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/37230. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 40 Appendix A: Provincial Tax Revenue, KP Province > > > F I G U R E A 1 - Provincial Tax Revenue, KP Province Source: KP Finance Department, Actual Receipts and Expenditure Report FY2020–21. Note: The figure represents the sum of direct and indirect provincial taxes. > > > TABLE A1 - Projected and actual tax revenue for selected provincial taxes, KP Province 2018–19 2019–20 2022–2021 Actual Actual Actual as % of as % of as % of Projected Actual projected Projected Actual projected Projected Actual projected Land Revenue 2.55 2.86 112% 3.05 2.44 80% 2.75 3.38 123% UIPT 1 1.26 126% 1.4 0.87 62% 2 1.62 81% Tax on Transfer of Property 0.27 0.22 81% 0.3 0.25 83% 0.31 0.07 23% Agricultural Income 0.11 0.08 73% 0.09 0.07 78% 0.09 0.09 100% Motor Vehicle Tax, R. Permit, Fitness 2 1.84 92% 1.85 1.33 72% 2.14 1.71 80% Source: KP Finance Department, Actual Receipts and Expenditure Report FY2020–21. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 41 Appendix B: Survey Sampling Strategy The target number of respondents for the taxpayer survey, > > > shown in Table B1—disaggregated by sex, tax, and location— Table B1 - Target number of respondents for the was established before the field work began. The purpose of taxpayer survey this Table was to provide quotas for different types of respon- Percent- dents who were likely to pay each tax in the various field tax of- Men Women Total age fices, and it guided the survey team’s efforts to identify respon- Peshawar 360 150 510 51% dents. This method is often referred to as Quota Sampling, UIPT 20 50 70 where the sampling is not random, and survey teams recruit Motor vehicle tax 200 60 260 respondents according to the set criteria until the established FARD 70 20 90 quotas are met (Stantcheva 2022). In this study, the sampling framework oversampled women to ensure a sufficient number Mutation 70 20 90 of female respondents. Mardan 160 150 310 31% UIPT 20 50 70 The target numbers were determined based on experience Motor vehicle tax 20 60 80 during the piloting of the survey and the following consid- FARD 60 20 80 erations. First, the sample size was reduced from 1,500 to Mutation 60 20 80 1,000 respondents because of the low response rate in the Abbottabad 60 40 100 10% pilot. Second, the target was to interview 60 percent of re- spondents by telephone and 40 percent in-person, as it was UIPT 20 20 40 expected that there would be fewer in-person taxpayers being Motor vehicle tax 40 20 60 interviewed. Third, given the difficulty of reaching female re- Kohat 40 40 80 8% spondents by telephone using the taxpayer database in the pi- FARD 20 20 40 lot stage, the target number for female in-person respondents Mutation 20 20 40 was increased. Fourth, given the lack of phone numbers for Total 620 380 1000 100% all UIPT taxpayers and Mardan MVT taxpayers, in-person in- terviews for UIPT taxpayers in Peshawar, Mardan, and Abbot- tabad and the MVT taxpayers in Mardan were seen as likely to be necessary. Fifth, for Peshawar, the target number of tele- phone interviews for male respondents paying MVT was set high because it was assumed that male MVT taxpayers also paid UIPT. Sixth, the target number of female respondents was set higher for MVT and UIPT than for FARD and mutation fees because it was anticipated that more women pay MVT or UIPT than FARD or mutation fees at the tax offices. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 42 Appendix C: Survey Methodology Gallup Pakistan piloted the survey instrument and conducted > > > the data collection for the taxpayer survey, focus group dis- Table C1: Survey sites for in-person interviews and cussions, and key informant interviews. The project director taxpayer databases for telephone interviews and staff from the IDA-funded Khyber Pakhtunkhwa Revenue In-person interviews at field Survey sites Mobilization and Public Resource Management Program tax offices (KPRMP) also participated and assisted in planning and im- Peshawar, Mardan, and plementing the study. FARD fee and mutation fee Kohat In-person interviews of taxpayers visiting field tax Motor vehicle registration and Peshawar, Mardan, and token tax, urban immovable offices Abbottabad property tax For the face-to-face interviews, the agreement with BOR and Telephone interviews from Districts ETNCD was to cover survey sites in (a) locations with heavy taxpayer databases traffic, given that more female taxpayers were likely to be Peshawar, Mardan, and found there, namely Peshawar and Mardan; and (b) one other FARD fee and mutation fee Kohat location, resulting in the survey sites in Table C1. Abbottabad Motor vehicle registration and Peshawar and Abbot- was chosen by the ETNCD because the Abbottabad TFC was token tax tabad considered a possible site for providing women-focused ser- vices. Kohat was chosen by BOR to represent a district in the South region. The BOR and ETNCD assigned focal persons in the field offices (at TFCs or SDCs) with whom the survey firm Telephone interviews of female clients of the coordinated on interview dates and logistics. These focal per- National Rural Support Program (NRSP) sons played a key role in reaching out to female respondents The National Rural Support Program is a not-for-profit orga- interviewed at these offices. All women visiting the tax offices nization that provides support, including training and microfi- were accompanied by male relatives. The focal persons also nance, to poor households through community organizations. provided office spaces so that female or male respondents The NRSP in KP province provided access to female clients of could be interviewed in private. the NRSP, who were reached by telephone. Telephone interviews of taxpayers in the provincial taxpayer databases The BOR and ETNCD provided access to the taxpayer’s data, which contained taxpayers’ phone numbers for the following taxes: the motor vehicle registration fee and token tax from Peshawar and Abbottabad, and FARD fee and mutation fee from Peshawar, Mardan, and Kohat. Phone numbers were not collected for all urban immovable property tax (UIPT) taxpay- ers or, in Mardan, for the motor vehicle tax. Therefore, indi- viduals in these data bases could not be reached by phone. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 43 Appendix D: Number of Study Participants > > > > > > Table D1: Number of in-person interviews, by district Table D3: Number of FGD and KII participants Men Women Number of Location Peshawar1 20 137 participants Tax facilitation center 0 18 Female Total 35 taxpayers Government or tax office 1 106 FGD 1 8 Peshawar Respondents’ homes or offices 19 5 FGD 2 9 Peshawar Other 0 8 FGD 3 9 Mardan Mardan 42 54 FGD 4 9 Mardan Tax facilitation center2 34 0 Female service Service delivery center 0 54 Total 15 providers (staff) Government or tax office 5 0 BOR 8 Peshawar Respondents’ homes or offices 3 0 ETO 7 Hayatabad Abbottabad 19 0 Key informant Tax facilitation center2 19 0 Total 2 interviews Kohat 0 46 Deputy Service delivery center 0 46 Director, 1 Peshawar Grand Total 81 237 BOR Source: Taxpayer Survey. Director Notes: 1 There are no SDCs in Peshawar. FARD/mutation taxpayers are serviced at government tax offices. Respondents for UIPT and FARD/mutation taxpay- General, 1 Peshawar ers in Peshawar were interviewed in government tax offices, challan offices, or respondents’ homes. ETNCD 2 Women were not available for interviews at TFCs in Abbottabad and Mar- Source: Taxpayer Survey. dan because they were not present or were reluctant to be interviewed. > > > Table D2: Number of telephone interviews from the taxpayer database, by district Men Women Peshawar 347 3 Motor vehicle tax 162 2 FARD 124 1 Mutation 71 0 Mardan 1 147 0 FARD 50 0 Mutation 97 0 Abbottabad 17 0 Motor vehicle tax 17 0 Kohat 157 1 FARD 136 1 Mutation 21 0 Total 678 4 Source: Taxpayer Survey. Notes: There were no telephone numbers for UIPT taxpayers. 1 Phone numbers for taxpayers of motor vehicle registration and token tax were available for Peshawar and Abbottabad, but not for Mardan. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 44 Appendix E: Individual Characteristics of Respondents Respondents in the taxpayer survey are more educated but this may partly be due to the taxpayer data omitting UIPT and have higher literacy rates than the provincial aver- taxpayers or Mardan’s motor vehicle taxpayers, who are ex- age, particularly so for women. Table E1 indicates that 85 pected to be more educated. percent of men are literate, which is 9 percentage points high- er than the male provincial average (76 percent); for women, Most women are not in paid employment (employers, 71 percent are literate, which is 31 percentage points higher waged employees, or self-employed), but there are differ- than the female provincial average (40 percent).31 It is likely ences by data source. A higher proportion of women in the that provincial taxpayers are more educated than the general NRSP list are self-employed (10 percent) compared to women population because they must be asset owners (of land, resi- interviewed in-person (1 percent) and are twice as likely to dential and commercial property, and motor vehicles) to be be employees (14 percent versus 7 percent of women inter- in the tax net, and wealthier individuals are expected to be viewed in-person) (Figure E2). The reverse is true for men: more educated. However, there is no census on taxpayers of men interviewed in person are more likely to be self-employed provincial taxes to be used to evaluate whether the survey is than men interviewed by phone (taxpayer database). representative of the provincial taxpayer population. > > > Men have higher education levels than women, but there Table E1: Individual and household characteristics of are also differences by data source. Figure E1 shows that survey respondents women in the NRSP client list (by phone interviews) are gen- (1) (2) (1)-(2) erally more educated than women interviewed in-person, with Men Women Difference a higher percentage of bachelor degree holders for NRSP Age 41.2 44.7 -3.5*** clients (31 percent) than for the latter group (19 percent). Married 91.2% 87.6% 3.6%* Yet, there is heterogeneity within these groups: women inter- Literate 85.4% 70.8% 14.6%*** viewed in person in Peshawar are also highly educated: 29 percent have a bachelor’s degree, and 58 percent graduated Number of 759 394 respondents from middle school or obtained SSCs/A levels. In contrast, Source: Authors’ calculations based on Taxpayer Survey. male in-person respondents are highly educated (41 percent Note: * <0.1, ***<0.01. with a bachelor’s degree) and are better educated than men interviewed by phone (28 percent with a bachelor’s degree), 31. The average literacy rates in KP province (excluding merged areas) are given in Pakistan Bureau of Statistics (2021). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 45 > > > F I G U R E E 1 - Completed educational attainment Source: Authors’ calculations based on Taxpayer Survey. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 46 > > > F I G U R E E 2 - Employment status Source: Authors’ calculations based on Taxpayer Survey. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 47 Appendix F: Reasons for Not Purchasing Goods Online Source: Authors’ calculations based on Taxpayer Survey. Note: The sample in this graph represents individuals who have not purchased anything online. *** 1% significant, ** 5% significant. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 48 Appendix G: Respondents’ Probability of Paying Provincial Taxes (conditional mean) (1) (2) (3) (4) (5) UIPT Motor vehicle FARD or Professional tax Paid tax on taxes mutation behalf of others Female -0.071*** -0.068** -0.146*** -0.076*** -0.090*** (0.021) (0.033) (0.029) (0.024) (0.019) R-squared 0.428 0.349 0.180 0.136 0.144 No. of observations 1,153 1,153 1,153 1,153 1,153 Source: Authors’ calculations based on Taxpayer Survey. Note: Coefficients represent estimates from linear probability models predicting that the respondent is responsible for tax payment (for all respondents). Robust standard errors are in parentheses. The regression also includes (but not shown here) age, educational attainment, whether respondent can read and write in English, household asset ownership, employment status, households’ source of drinking water, mobile ownership, presence of children in household, households with no male adults, rural area, district of residence, whether interview was conducted in Pashto or Hindko, and whether the respondent was interviewed in person. *** 1% significant, ** 5% significant, * 10% significant. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 49 Appendix H: Taxpayers’ Methods Used for Provincial Tax Payment or Property Registration Taxpayers only (1) (2) (3) (4) (5) (6) TFC SDC Bank app Banks Post office Used agent Female -0.056 0.105 -0.082 -0.345*** -0.068 0.058 (0.076) (0.068) (0.052) (0.080) (0.053) (0.055) R-squared 0.184 0.259 0.130 0.238 0.208 0.127 N of observation 579 579 579 579 579 579 Source: Authors’ calculations based on Taxpayer Survey. Note: The coefficients represent linear probability models predicting use of methods for tax payment or property registration among taxpayers. Robust standard errors are in parentheses. The regression also includes (but not shown here) age, educational attainment, whether respondent can read and write in English, household asset ownership, employment status, households’ source of drinking water, mobile ownership, presence of children in household, households with no male adults, rural area, district of residence, whether interview was conducted in Pashto or Hindko, and whether the respondent was interviewed in person. *** 1% significant, ** 5% significant, * 10% significant. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 50 Appendix I: Use of Social Media as a Source of Tax Information by Age Source: Authors’ calculations based on Taxpayer Survey. Note: The graphs show the estimated probability of the important source of tax information from a linear probability model by education. The vertical lines show the 95% con- fidence interval. The regression also includes (but not shown here) age, educational attainment, whether respondent can read and write in English, employment status, households’ source of drinking water, mobile ownership, presence of children in household, households with no male adults, rural area, district of residence, whether interview was conducted in Pashto or Hindko, and whether the respondent was interviewed in person. GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 51 Appendix J: Suggested Timeline of Recommended Action Plan Recommended action Short term Medium term Include washrooms and areas for women in the new Women’s washroom and areas for infrastructure investments planned to establish one- women established in 12 TFCs. stop TFCs. Establish service counters dedicated to serving Dedicated service counters for women in TFCs. women established. 32% of services provided to women.32 Review the institutional structure and roles, sustained Institutional structure and roles training, and means of achieving a safe and equitable reviewed. work environment—including equitable promotions, flexible work arrangements, anti-harassment policies Training programs and promotions and practices, and physical safety—to recruit and ensure equitable opportunities for retain women. women. Safe and equitable work environment conditions reviewed and instituted. Review implications of computerization of land Property processes for women records and digitization of vehicle registration and reviewed and simplified. establishment of service counters for women on property registration processes. Use easy-to-understand terminology in Urdu or Pashto Terminologies in guidance for guidance on tax payments and registration forms. simplified. Tailor tax campaigns or information strategies to reach Communication campaigns different taxpayers. conducted twice a year 1. Conduct field-level education campaign efforts, to include these tailored such as TV or radio programs, and involve NGOs programs.33 or field staff in disseminating information on wom- en’s property rights and tax compliance to reach less educated women and men. 2. Use various social media channels to reach out to younger and more educated taxpayers. 3. Target sensitization campaigns to women but also to men who could play a role in assisting women with property registration. Compile, digitize, and sex-disaggregate taxpayer and KP tax authorities’ data integration service user data to help develop tailored taxpayer is sex-disaggregated (with programs and services that meet the needs of women taxpayer CNIC/Name).34 and men. Service user data is sex- disaggregated. 32. The target of 32 percent is consistent with the Results Framework Matrix in Annex 1 of the KPRMP (World Bank 2019b, 16). 33. The KPRMP’s program action plan includes communication campaigns to be conducted twice a year (World Bank 2019b, 41). 34. This target is consistent with the Results Framework Matrix in Annex 1 of the KPRMP (World Bank 2019b, 14). GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 52 GENDER AND TAXPAYER STUDY IN KHYBER PAKHTUNKHWA, PAKISTAN <<< 53