1# RECORD e8f48e7e91bb043c222f2cfbe026c67e0e422a5f2# TITLE: Valuing Neighbourhood Soundscapes in Urban Housing Markets: Evidence from Karachi3# YEAR: 2026.0 VENUE: Pakistan Development Review DOI: 10.30541/v64i3pp.241-2614# AUTHORS: Ghamz E. Ali Siyal; Heman Das Lohano; Johane Dikgang; Ayesha Zia5# PAGES: 21 CANDIDATE TABLE PAGES (1-based): [11, 12, 16, 17]6# ABSTRACT: Noise pollution in Karachi has already exceeded WHO-recommended environmental thresholds. It causes sleep disturbances, hearing impairment, stress, and anxiety. However, there is a lack of evidence on how these effects transferred into housing market. This study extends a conceptual framework and empirically explores household data on perceived neighbourhood soundscapes and self-reported housing values. We estimate hedonic pricing models to examine how the presence of natural and non-natural soundscapes around respondent’s residence affects housing values. This study conducted a primary data survey of 507 respondents affiliated with the Institute of Business Administration (IBA), Karachi. Using hedonic regression results, this study is the first to confirm the effect of natural and non-natural soundscapes on housing values, along with structural, locational, and neighbourhood attributes. The results recommend designing and installing soundproof windows in homes and developing better urban planning and management policies to ensure a good quality of life in Karachi. 789===== PAGE 1 (score 9, numeric-paren 0, table-words 0, var-words 9) =====1011© The Pakistan Development Review 1264:3 (2025) pp. 241–261 13DOI: 10.30541/v64i3pp.241-261 14Published online: 24 July,2026 15 16Valuing Neighbourhood Soundscapes in Urban Housing 17Markets: Evidence from Karachi 18GHAMZ E ALI SIYAL1, HEMAN DAS LOHANO, 19 JOHANE DIKGANG, and AYESHA ZIA 20Received: 16 October 2025/ Accepted: 23 June 2026 21 22 23Noise pollution in Karachi has already exceeded WHO-recommended environmental 24thresholds. It causes sleep disturbances, hearing impairment, stress, and anxiety. However, there is a 25lack of evidence on how these effects transferred into housing market. This study extends a 26conceptual framework and empirically explores household data on perceived neighbourhood 27soundscapes and self-reported housing values. We estimate hedonic pricing models to examine how 28the presence of natural and non-natural soundscapes around respondent’s residence affects housing 29values. This study conducted a primary data survey of 507 respondents affiliated with the Institute 30of Business Administration (IBA), Karachi. Using hedonic regression results, this study is the first 31to confirm the effect of natural and non-natural soundscapes on housing values, along with structural, 32locational, and neighbourhood attributes. The results recommend designing and installing 33soundproof windows in homes and developing better urban planning and management policies to 34ensure a good quality of life in Karachi. 35JEL Classification: R21, R31, R23, Q53, I18, Q51 36Keywords: Natural Soundscapes, Non-natural Soundscapes, Noise Pollution, 37Proximity to Amenities, Neighbourhood Characteristics, Housing 38Market Values. 391. INTRODUCTION 40Megacities in Pakistan, for instance Karachi, have witnessed environmental 41degradation such as reduction of green spaces as a negative externality due to unplanned 42and unmanaged urbanisation (Asian Development Bank, 2024 Mehdi, et al. 2011, 43Qureshi, et al. 2010). The presence of green spaces is well-documented to provide benefits 44to human health and mental well-being via interaction with nature (Alvarsson, et al. 2010, 45Hedblom, et al. 2014, Uebel, et al. 2022). The possible pathways of such interactions are 46based on visual stimuli and soundscapes that are responsible for improving human health 47and mental well-being (Ratcliffe 2013, Smalley, et al. 2022). In particular, bird sounds are 48 491 Ghamz E Ali Siyal < gasiyal@iba.edu.pk> is Assistant Professor, Institute of Business Administration 50(IBA), Karachi, and Former Postdoctoral Researcher, Department of Economics and Finance/The Water School, 51Florida Gulf Coast University (FGCU), Fort Myers, Zip code 33967, Florida, USA. Heman Das Lohano 52< hlohano@iba.edu.pk> is Professor, School of Economics and Social Sciences (SESS), Institute of Business 53Administration (IBA), Karachi and Visiting Professor, Department of Applied Economics, University of 54Minnesota. Johane Dikgang < jdikgang@fgcu.edu> is Associate Professor, Department of Economics and 55Finance/The Water School, Florida Gulf Coast University (FGCU), Fort Myers, Zip code 33967, Florida, USA. 56Ayesha Zia < ayeshazia@iba.edu.pk> is Assistant Professor, School of Economics and Social Sciences (SESS), 57Institute of Business Administration (IBA), Karachi. 58Authors’ Note: We are grateful to acknowledge the financial support provided by the Competitive 59Research Grant (CRG) of the Institute of Business Administration (IBA), Karachi. The authors also sincerely 60thank Mr. Muhammad Nawaz, Mr. Abdul Qadeer, Mr. Dolat Kumar, and Mr. Muhammad Saad for volunteering 61their assistance in supervising data collection while serving as the graduate students at the IBA, Karachi. 626364===== PAGE 11 (score 60, numeric-paren 40, table-words 3, var-words 10) =====6566Valuing Soundscapes and Neighborhood Quality in Urban Housing Markets 67251 68The identification strategy used in this study is a regression model. For robustness 69of results, we used two models which have only slight difference in structural attributes. 70Both models follow empirical Equation 2, but only the number of restrooms is added in 71place of the number of bedrooms. The details of results from both models are given in the 72next section. 736. RESULTS 74In this section, the regression model includes a large number of explanatory 75variables compared to the sample size for capturing a wide range of housing, 76neighbourhood, and environment characteristics given in Table 1. Despite rich number of 77variables, it is possible that our analysis may have potential omitted variables and other 78sources of bias in our estimated coefficients. Therefore, results analysis should be 79interpreted as associations rather than causal effects. Secondly, these models’ results are 80based on housing value perceptions among the respondents affiliated with the Institute of 81Business Administration (IBA) that may not accurately reflect urban population of 82Karachi. For multicollinearity, we found moderate or low multicollinearity among 83variables for both models. Results of Variance Inflation Factor (VIF) for both models are 84given in the Appendix 2. Full regression results of both models are given in the Appendix 851 and abridged results are given in the following Table 2. 86Table 2 87Regression Results for Hedonic Model 88 89(1) 90(2) 91Variables 92Log (house value) 93Log (house value) 94Structural Attributes 95Number of Bedrooms 960.19084*** 97 98 99(0.01905) 100 101Number of Restrooms 102 1030.22406*** 104 105 106(0.01556) 107Solar 1080.19034** 1090.15393* 110 111(0.08700) 112(0.07881) 113Gas 1140.27690*** 1150.25384*** 116 117(0.10152) 118(0.09281) 119Water Supply 1200.30605*** 1210.21211*** 122 123(0.08358) 124(0.07702) 125Locational Attributes 126Commercial 127-0.32337*** 128-0.20353*** 129 130(0.07240) 131(0.06739) 132Grocery Store within 1 km 1330.17134** 1340.12619* 135 136(0.07043) 137(0.06453) 138School within 1 km 139-0.19937*** 140-0.22155*** 141 142(0.07551) 143(0.06892) 144Police Station within 1 km 1450.02213 1460.14729* 147 148(0.09772) 149(0.08870) 150Environmental Attributes 151Availability of Park within 1 km 1520.34378*** 1530.29760*** 154 155(0.06762) 156(0.06208) 157Neighbourhood & Soundscape Attributes 158Friendly Neighbourhood 159-0.14998* 160-0.10355 161 162(0.08965) 163(0.08213) 164Hear birds often 165-0.22090** 166-0.16602** 167168169===== PAGE 12 (score 28, numeric-paren 12, table-words 5, var-words 5) =====170171252 172 173 174Siyal, Lohano, Dikgang & Zia 175 176(0.08997) 177(0.08251) 178Chowks 179-0.12687* 180-0.11883* 181 182(0.06910) 183(0.06324) 184Constant 18515.71325*** 18615.64518*** 187 188(0.18442) 189(0.16732) 190Observations 191484 192484 193R-squared 1940.4572 1950.5498 196Note: Full regression results with insignificant variables of both models are available in Appendix 1. 197 198Regression results from Hedonic models given in Table 2 are not causal effects but 199show positive or negative association with reported housing values by respondents. Table 2002 provide insightful findings related to structural attributes, locational attributes, 201neighbourhood attributes, environmental attributes, soundscape attributes. Our results 202from both models particularly agree with most of previous literature such as Ahmed, et 203al. (2020), Wahid, et al. (2023) and other prior hedonic studies conducted in other cities 204of Pakistan confirming structural, locational, neighbourhood attributes effects on housing 205values. The important contribution of this study lies in investigating the relationship of 206new attributes from environmental, neighbourhood, and soundscape attributes to housing 207values. However, the coefficients are found to be higher than previous studies. The R-208square from second models is higher than model 1. It means that explanatory variable 209“number of restrooms” in second model explain variation in housing values better than 210“number of bedrooms” in model 1. Presence of more restrooms in a house increases 211convenience for joint families that could be a possible reason for higher R-square values. 212 2136.1. Structural Attributes 214Structural attributes increase housing prices significantly. They directly affect a 215property’s functionality, convenience, and appeal to potential buyers. For instance, access 216to utilities of gas and water significantly increases house values by 25-27 percent and 21-21730 percent respectively. Considering other structural attributes, the number of bedrooms 218and restrooms in a house increases house size which positively correlates with house 219values. We considered both separately because conceptual relationship between both 220could potentially bias the results. Results describe that with the increase of one additional 221bedroom and restroom housing values increase by 19 percent and 22 percent, respectively. 222Comparing model 1 and 2, the key difference is number of bedroom variables in 223model 1 and number of restrooms is added in model 2. The model 2 has higher R-squared 224value (0.54) than model 1 (0.46) which suggests that the higher number of restrooms 225explains a greater proportion in housing values within the sampled households. One of 226potential reasons behind this is number of bedrooms usually varies less across houses of 227similar size, but the presence of additional restrooms may indicate higher household 228convenience especially in joint family system. 229This makes quite sense as there are differences in rent and housing values in the 230actual market if we compare, for instance, two bedrooms with a three-bedroom house or 231an apartment with an additional restroom. Along with that, houses which have access to 232solar systems have 15 to 19 percent higher property values than who don’t have it. In this 233case, solar acts as an asset and it reflects some income effect. People with solar systems 234are living in houses with significant differences in property values as compared to people 235who don’t have solar supported electricity supply. Therefore, having a solar system in a 236house has higher value compared to those who don’t have it. 237 238 239240241===== PAGE 13 (score 13, numeric-paren 0, table-words 0, var-words 13) =====242243Valuing Soundscapes and Neighborhood Quality in Urban Housing Markets 244253 2456.2. Locational Attributes 246Locational attributes are also related to housing values as mentioned in the literature. 247We found only three important factors from locational attributes which affect housing values. 248Firstly, residential type affects housing values. For example, housing values decrease by 20 to 24933 percent if these are in a commercial area as compared to residential areas. Homes located 250in commercial areas are generally less desirable for residential purposes due to noise, pollution, 251and a lack of privacy. In a city like Karachi, a house situated in a busy commercial area like 252Saddar might be less appealing to families looking for a quiet residential environment. In 253contrast, homes in purely residential areas like Gulshan-e-Iqbal or DHA tend to be more 254valuable, as they offer a more peaceful and family-friendly atmosphere.Secondly, the presence 255of a school near to respondent’s house reduces its value. Results show that houses that have 256access to schools within 1 km have 19 to 22 percent lower housing values as compared to 257houses far from schools. It is a surprising result but a potential reason for that can be associated 258with traffic jams that increase delays while commuting. Similar results are found in Ahmed, 259et al. 2020, where presence of higher number of schools and hospitals reduces housing values 260in nearby surroundings by 9 percent and 13 percent, respectively. The potential reason 261provided by Ahmed, et al. 2020, is congestion and overcrowded areas. Therefore, our results 262show negative school proximity effect, ranging from –19 to –22 percent, is significantly larger 263than most other proximity variables in the model. This indicates that the congestion externality 264has a substantial economic impact rather than being marginal. Thirdly, unlike schools, 265locational amenities such as grocery stores and police station have longer and flexible 266operating hours which can be potential reason for less traffic congestion and positive effects 267on housing values. 268Similarly, access to grocery stores is also an important factor that affects housing 269values. Results found that people who live near grocery stores, i.e., within 1 Km, have 12 to 27017 percent higher housing values than their counterparts, i.e., who live more than 1 Km away. 271Its potential reason can be people consider access to nearby grocery stores so that it saves the 272time required in fulfilling daily necessities. In addition, access to police station within 1 km 273increases house values by 14 percent as compared to houses away from 1 km in model 2. 2746.3. Environmental Attributes 275Results from the environmental attributes also describe unique relationship with 276housing values. Results show that housing values increase by 29 to 34 percent if people 277have access to a park within 1 km of their house. Many people love to stay in a peaceful 278environment, and parks are the best place for this. Living closer to a park enhances the 279quality of life by offering residents a place to relax, exercise, and enjoy nature, making 280nearby properties more desirable. However, environmental pollution related to frequent 281choking of sewage, open air dumping, and open air burning of trash does not have 282significant effect on housing values. One potential reason for insignificant results is that 283the sample of respondents does not find such problems in their locality. Secondly, we 284have a sample of respondents that are affiliated with IBA institution that may not fully 285represent households across Karachi. The sample may also have similar income levels. 286Thirdly, the effect of these disamenities such as open-air dumping or burning of trash may 287affect housing values if respondent is living next to it. Respondents living very close to 288these sites may have higher exposure to negative visual stimuli by watching such sites or 289odor smells have effect housing values. However, our sample may not have respondents 290very close or next to these sites. 291292293===== PAGE 16 (score 79, numeric-paren 55, table-words 1, var-words 15) =====294295256 296 297 298Siyal, Lohano, Dikgang & Zia 299APPENDIX 300 301Table 3 302Appendix 1: Full Regression Results Model 1 and 2 303Variables 304(1) 305(2) 306Log (house value) 307Log (house value) 308Structural Attributes 309Number of Bedrooms 3100.19084*** 311 312 313(0.01905) 314 315Number of Restrooms 316 3170.22406*** 318 319 320(0.01556) 321Solar 3220.19034** 3230.15393* 324 325(0.08700) 326(0.07881) 327Gas 3280.27690*** 3290.25384*** 330 331(0.10152) 332(0.09281) 333Water Supply 3340.30605*** 3350.21211*** 336 337(0.08358) 338(0.07702) 339Locational Attributes 340Commercial 341-0.32337*** 342-0.20353*** 343 344(0.07240) 345(0.06739) 346Grocery Store within 1 km 3470.17134** 3480.12619* 349 350(0.07043) 351(0.06453) 352School within 1 km 353-0.19937*** 354-0.22155*** 355 356(0.07551) 357(0.06892) 358College within 1 km 3590.17704 3600.10853 361 362(0.11147) 363(0.10220) 364University within 1 km 365-0.06739 366-0.04698 367 368(0.13293) 369(0.12162) 370Hospital within 1 km 3710.08957 3720.04538 373 374(0.08688) 375(0.07950) 376Bus stop within1 km 377-0.06862 378-0.03853 379 380(0.07352) 381(0.06703) 382Police Station within 1 km 3830.02213 3840.14729* 385 386(0.09772) 387(0.08870) 388Railway Station within 1 km 389-0.22920 390-0.28402 391 392(0.27549) 393(0.25189) 394Airport within 1 km 395-0.08662 396-0.17339 397 398(0.36110) 399(0.33029) 400Post office within 1 km 4010.04187 4020.02894 403 404(0.11913) 405(0.10902) 406Bank within 1 km 4070.04841 4080.02741 409 410(0.06972) 411(0.06369) 412Environmental Attributes 413Trees 4140.09482 4150.06448 416 417(0.07637) 418(0.06997) 419Availability of Park within 1 km 4200.34378*** 4210.29760*** 422 423(0.06762) 424(0.06208) 425Open Air Waste Dump within 1 km 4260.02968 4270.03656 428 429(0.07530) 430(0.06888) 431Open Air Burning of Waste 432-0.10674 433-0.07552 434435436===== PAGE 17 (score 58, numeric-paren 35, table-words 3, var-words 10) =====437438Valuing Soundscapes and Neighborhood Quality in Urban Housing Markets 439257 440 441(0.06896) 442(0.06306) 443Choked Sewer pipes 444-0.04820 445-0.05673 446 447(0.06954) 448(0.06356) 449Neighbourhood & Soundscape Attributes 450Poor roads 4510.02837 4520.02430 453 454(0.07313) 455(0.06693) 456Streetlight 457-0.02707 458-0.00476 459 460(0.08168) 461(0.07479) 462No garbage Collection 463-0.17426 464-0.17237 465 466(0.23352) 467(0.21370) 468Crime 469-0.11143 470-0.09445 471 472(0.06979) 473(0.06385) 474Respect Neighbourhood 4750.12975 4760.12689 477 478(0.08700) 479(0.07959) 480Trust Neighbourhood 4810.02358 482-0.00760 483 484(0.09925) 485(0.09089) 486Friendly Neighbourhood 487-0.14998* 488-0.10355 489 490(0.08965) 491(0.08213) 492Hear Birds often 493-0.22090** 494-0.16602** 495 496(0.08997) 497(0.08251) 498Hear Traffic often 4990.03630 5000.02107 501 502(0.06815) 503(0.06227) 504Hear Construction often 5050.04938 5060.03640 507 508(0.06699) 509(0.06132) 510Chowks 511-0.12687* 512-0.11883* 513 514(0.06910) 515(0.06324) 516Constant 51715.71325*** 51815.64518*** 519 520(0.18442) 521(0.16732) 522Observations 523484 524484 525R-squared 5260.4625 5270.5498 528 529Table 4 530Appendix 2: Model 1 Variance Inflation Factor (VIF) 531 532VIF 5331/VIF 534Trust Neighbourhood 5351.599 536.625 537Friendly Neighbourhood 5381.576 539.635 540Respect Neighbourhood 5411.524 542.656 543Distance Rail within 1 km 5441.448 545.691 546Distance Airport within 1 km 5471.394 548.717 549Distance School within 1 km 5501.393 551.718 552Water Supply 5531.369 554.73 555Distance College within 1 km 5561.35 557.741 558Gas 5591.349 560.741 561No Sweeping 5621.339 563.747 564Poor Road 5651.331 566.751 567Distance Bus Stop within 1 km 5681.329 569.753 570Distance Post Office within 1 km 5711.326 572.754 573Distance Bank 1 km 5741.27 575.788 576Number of Bedrooms 5771.269 578.788 579Distance Hospital within 1 km 5801.265 581.791 582Distance Police within 1 km 5831.264 584.791 585586587===== PAGE 18 (score 18, numeric-paren 0, table-words 1, var-words 12) =====588589258 590 591 592Siyal, Lohano, Dikgang & Zia 593Distance Kachra 1 km 5941.263 595.792 596Witness Burning 5971.24 598.807 599Witness Choked pipes 6001.237 601.808 602Distance Grocery within 1 km 6031.234 604.81 605Streetlight 6061.233 607.811 608Hear Traffic Frequently 6091.194 610.837 611Distance Park within 1 km 6121.192 613.839 614Noisy Chowks 6151.188 616.842 617Crime 6181.166 619.857 620Solar 6211.156 622.865 623No Garbage Collection 6241.154 625.867 626Distance University within 1 km 6271.141 628.877 629Hear Construction Frequently 6301.134 631.882 632Trees 6331.131 634.884 635Commercial 6361.127 637.888 638Hear Birds Frequently 6391.096 640.912 641Mean VIF 6421.281 643. 644 645Table 5 646Appendix 2: Model 2 Variance Inflation factor (VIF) 647 648VIF 6491/VIF 650Trust Neighbourhood 6511.601 652.625 653Friendly Neighbourhood 6541.579 655.633 656Respect Neighbourhood 6571.522 658.657 659Distance Rail within 1km 6601.445 661.692 662Distance Airport within 1km 6631.392 664.718 665Water Supply 6661.388 667.72 668Distance School within 1km 6691.385 670.722 671Distance College within 1km 6721.355 673.738 674Gas 6751.346 676.743 677No Sweeping 6781.338 679.747 680Poor Road 6811.331 682.751 683Distance Post Office within 1km 6841.325 685.754 686Distance Bus Stop within 1km 6871.319 688.758 689Number of Rest Room or Washrooms 6901.308 691.764 692Distance Bank within 1km 6931.265 694.79 695Distance Hospital within 1km 6961.264 697.791 698Distance Kachra within 1km 6991.262 700.793 701Distance Police within 1km 7021.244 703.804 704Witness Burning 7051.238 706.808 707Distance Grocery within 1km 7081.237 709.808 710Witness Choked Pipes 7111.234 712.81 713Streetlight 7141.234 715.81 716Distance Park within 1km 7171.2 718.834 719Hear Traffic Frequently 7201.19 721.84 722Noisy Chowks 7231.188 724.842 725Crime 7261.166 727.858 728Commercial 7291.165 730.858 731No Garbage Collection 7321.153 733.867 734