Analysing the Housing Landscape In Major Cities of Pakistan
Background
Introduction
Prices for housing vary from cities to towns and villages. The metrics for calculated the price of 2-bedroom apartment may depend on the locality, as well as if the apartment is furnished or not. Properties are not always for sale as well; people tend to rent properties as a source of extra income. Prices also depend on if the property has been declared as commercial land or residential. The focus of this report is to look at the housing landscape of Pakistan and some of the factors that may affect price in different cities. Pakistan faces a shortage of decent and affordable housing, and the real estate market is not properly regulated by the government (Delmendo 2019). As the new government took over in 2018, several reforms were introduced to counteract the rising inflation in the country which caused the prices of property to increase per square feet (different for different cities) but when adjusted for inflation the prices were seen to decrease overall.
Sigma Properties (leading real estate company in Pakistan) stated the real estate market accounts for 70–75% of the country’s wealth and it can be valued at over $400 billion. In 2017, Pakistan spent over $5.2 billion every year on infrastructure with prices for apartments rising by up-to 120% in 2018 (Akhtar et al. 2017). Property has been regarded as a good income source, where you can easily buy a commodity and collect rent, or you could just wait till the market regains strength and sell your asset for a higher price. However, the price of a property is affected by multiple factors, and it is important that all factors are taken into consideration before a purchase is made. Buying a house is not cheap, it is a lifetime commitment. The neighbourhood matters, the access to utilities, the number of bedrooms, the security etc. all matter. Zaman et al. (2021) stated that an increase in the prices of real estate is directly related to the income of consumers and in turn meaning the more the income the better the location, the security, and the status of a person.
Problem Statement
As stated in the introduction, many factors affect the price of a potential home, especially in a country like Pakistan where no regulations exist to keep the prices in check. The real estate market is said to be investor-centric meaning that the investor controls the prices. They buy properties in bulk and refuse to sell until the prices rise. Pakistan also has a non-existent mortgage sector due to high interest rates despite efforts of the State Bank to urge private banks to lend more. Althobaiti, et al., (2021) explained that prices are susceptible to their surroundings meaning if an impoverished neighbourhood is next to a wealthy one, the prices of the impoverished neighbourhood will rise. There is also a lack of reliable data in Pakistan that has not yet allowed a proper forecasting or predictive model to be built that curtails the essence of the real estate sector.
For example, documents are not readily available for private societies and some societies are not yet sanctioned by the development authority even though there are people living there. It is important to note that Pakistan is a developing economy, an even a little ripple can have everlasting results on the country. Property is one of the most lucrative sectors in the country and there are several policies being introduced to help boost the economy. However, there has been a lack of understanding of factors affecting the price of housing in Pakistan. Due to this lack of understanding, this report aims to investigate pricing of properties based on the available features to gain insight on how the market is currently operating.
Objectives
The objectives of this article can be stated as follows:
- How size of a house affects the price city-to-city?
- What are the preferences of people for housing i.e., house or apartment etc.?
- Exploring the ad type i.e., for sale or for rent is more common and which is more profitable.
Data Story
Housing Preferences
Among the three regions (provinces of Pakistan) only three were selected but only Punjab has the most cities. A basic summary that can be taken from the visualisation is that mostly houses, and flats are in high demand in all the cities while other options are non-existent or not attractive commodities. Lahore has houses as the most popular category followed by flats which is only 4% of the total houses in the city. Karachi, the first capital of Pakistan, has a much more even distribution between housing and flats, with flats just edging housing by a considerably smaller amount. Islamabad and Rawalpindi are the twin cities of Pakistan with only a few kilometres separating Rawalpindi from the capital Islamabad. The comparison between the two cities shows that people tend to live in housing and flats more in the capital as compared to its twin city. Based on the figure, we can safely say that the city with the highest number of listings is Karachi, with housing and flats alone contributing to more than 50,000 listings. This is followed by Lahore (less than 40,000 for housing and flats), then Islamabad (less than 30,000), Rawalpindi (less than 20,000) and Faisalabad (less than 9000 in total property listings). The reason for this is that Karachi and Lahore have the most industries in the country and these are also the most populated cities of Pakistan. Karachi is known for its high-rises and coastal views hence the larger number of flats than housing. Islamabad and Rawalpindi being neighbours share a much more smaller population breakdown than Lahore and Karachi and due to the shorter distance between the cities, people find it easier to commute between the cities even though according to the figure, more listings exist for Islamabad.
According to Delmendo (2019), the adjusted inflation in Pakistan causes house prices in Pakistan to decline. As in Lahore, the average housing price was PKR 10,402 (US$73) per square foot, up 6.25 percent year on year but down 2.89 percent when adjusted for inflation. Another example in Karachi, housing prices in Karachi averaged PKR 13,158 (US$93) per square foot, up 4.25 percent year on year but down 4.62 percent adjusted for inflation. In addition, in Islamabad, house pricing was PKR 9,985 (US$70) per square foot, up 7.01 percent year on year but down 2.2 percent in actual terms. Suppose the results of the data set analysis are related to the statement. In that case, it can be concluded that the high number of “House” type properties was due to adjusted inflation in 2019 in Pakistan which caused the prices of “House” to fall. This causes some people to buy a house at a lower price, causing a spike in the number of houses in 2019.
For Sale vs For Rent
People tend to rent property if they cannot afford the house or the flat. But rent like the price of property are affected by different factors. In a country like Pakistan, the lack of mortgage facilities, it is more favourable to live on rent than buy. (Delmendo, 2019). Dixon (2019) stated factors include the properties’ value in the market, condition of the properties, the location where the properties reside, and the local sales price of the properties. However, he also stated that there are financial risks to consider when converting properties into investment by renting it out.
To minimise these risks, the properties’ owner is advised to purchase an insurance, that would be an additional expense to the maintenance cost and taxes which the properties’ owner is required to cover. To ensure a fair comparison, factors such as the location of the properties, the type of properties, and the size of properties are taken into considerations. Moreover, as stated by Wiebe (2021), the properties’ market value is approximately like the sales price of the properties. Hence, in this case, sales price of the property is used as the representative of the properties’ market value.
According to Dixon (2019), for the properties to be profitable but still affordable, the rental price should range between 0.8% to 1.1% of the properties’ value. As can be observed in figure above, the percentage of the average rental price to average sales price did not fall without this range (the yellow highlighted zone). Majority are below this range except for ‘Lower Portion’ in the Islamabad Capital, this means that on average the rental properties do not yield as much profit as it should. As for the property type of ‘Lower Portion’ in the Islamabad Capital, on average this type of property would be able to yield a much higher percentage of profit for properties’ owners in Islamabad, with the rental price being 1.884% of the properties’ value.
Moving on with the analysis, it is important to note that prices for both rental and sale properties depend on different factors. The below figures aim to explain this trend properties in Pakistan.
As can be observed in figures above, there is an exponentially negative relationship between the size of the properties and the percentage of the average rental price to the average sales price. Dixon (2019) explains that when the properties are larger, they tend to have a higher market value, therefore, it is more marketable for larger properties to have a lower rental price in order to maintain the demand by ensuring that it is at an affordable price. However, from the properties’ owners’ perspective, this could be an issue, as larger properties would require higher maintenance cost and more expensive insurance. This would further lower the profit margin of the rental properties. However, especially with the exponential relationship observed in those figures, smaller properties would be more profitable in comparison to larger properties. And it is also important to observe inflation effects as pointed out in previous section when coming to terms with selling or renting any property. (Nguyen, 2020).
As can be observed from both figures above, with p-value of 0.0005229 which is less than 0.05, it can be concluded that at 95% confidence interval, there is enough evidence to suggest that there is a significant positive correlation between the average sales price and the date which the advertisement is added. However, the same cannot be said for the average rental price.
As can be observed from figures above, there is a slight positive correlation between the average rental price and the date which the advertisement is added, however, the summary table of the trend line in Figure 8 shows a p-value of 0.289278, which is more than 0.05, suggesting that at 95% confidence interval, there is insufficient evident to suggest that the correlation is significant. This means that the rental price is not as sensitive to time, hence, the economy as the sales price is, hence, the properties’ owners could use this factor in determining whether it would be more profitable to sell their properties during certain period or be more profitable to rent out their properties instead.
Prices per Square Feet
The size of the house also affects the price. The bigger the covered area the greater a person must pay for rent or to buy the entire property. The metrics used in the dataset were according to the terms that usually a property is measured in Pakistan but for ease of reference it was converted into square meters. The mode of calculation used is as follows:
- 1marla = 272.251 square feet
- 5marla = 1361.25 square feet
- 10marla = 2722.51 square feet
1kanal = 20marla = 5445 square feet
The table above shows the different groups that were created to check the average prices of properties based on their size. There are a total of 11 groups that were created to make the visualization below. The average house sizes in Pakistan range from 5marla to 1kanal and it also can go higher. That is why the breakdown in table above encompasses all the size metrics that were seen in the data.
Based on the bar chart, house area 0 to 1000 square meter in Lahore has the highest average price between other cities with the average price of RS 20,826,000 and Faisalabad is the lowest average price with RS 7,331,000. Karachi has the highest average price for both 1001 to 2000 square meters and 2001 to 3000 square meters with the average price of RS 118,072,000 and RS 197,202,000. As for the area of 3001 to 4000 square meter, Rawalpindi and Lahore may consider have a high average price with the differences by RS 14,672,000 while other cities have a significant difference between Rawalpindi and Lahore. From the chart, we can also see that Lahore have quite a high average price in most housing area size than other cities. For example, for the house size of 4001 to 5000, 6001 to 7000, 7001 to 8000, 9001 to 10000, and house area size which are more than 10,000 square meters. However, it is clearly shown that the average house price is not fixed based on the size of the house.
As a result, the price of the house does not really rely on the size of the house where smaller house size may cost more than a bigger size. For example, in Karachi city, house with size area of 0 to 1000 square meters have the average price of RS 17,054,000 and house with size area 5001 to 6000 square meters have the average price with only RS 19,178,000. These two big differences of house size do not have a significant effect towards the average price. However, this may conclude other external factors that have been mentioned by Zawadzki, et al. (2017) where housing area which are closer than a local landmark may cost more such as housing area near a historical landmark. Hussain, et al. (2019) also mentioned that housing area which are located near slums area in Pakistan may cause a negative impact towards the price.
Recommendations
The data story has shown how certain trends that are prevalent in the housing market of Pakistan. We have seen inflation affect the prices differently in different cities and we have also seen the preference of people when choosing between buying a house or renting a flat extra. Most of the data is dominated by listings between housing and flats with only Karachi sharing an almost even distribution between both types. We can summarize our recommendations as follows:
- As the need for high rise apartments increases, more marketing can be targeted in cities like Lahore, Islamabad, and Rawalpindi as they show potential to grow.
- Zameen.com can expand operations to Faisalabad to utilize the real estate potential as government policies become more construction friendly.
- Upper and Lower portion see good numbers where property prices are the highest (Karachi and Islamabad). This means an increase in marketing for these areas for maximum outreach.
- If the property type is ‘Lower Portion’ and the properties are located in Islamabad, it is more profitable for properties’ owner to convert it into investments by renting them out. Otherwise, it is more profitable to resell the properties.
- It is recommended for properties of smaller sizes to be rented out in comparison to that of large sizes, as smaller size properties can receive higher profit margin. For larger size properties, it would be more profitable to resell rather than renting it out.
- As sales price is sensitive to the time, hence, the economy’s situation, it is advisable for properties’ owner to sell their properties during the period where the price of the property is very high. Otherwise, when the properties’ values are lower, it is more profitable to rent out the properties, as the rental price is not time sensitive, hence, does not depend on the economy as much as the sales price.
- Hussain et al. (2017) showed how prices are affected by societies near slums. Urban development projects like the one being done by the government called Naya Pakistan Housing Scheme can help aid in cleaning up the area and providing low-income affordable housing to people.
References
- Akhtar, M., Rao, Z.-u-R., Shafiq, M., & Umar, M. (2019 December 19). Impact of Monetary Policy on House Prices: Case of Pakistan. International Journal of Housing Markets and Analysis, Vol 13 N0. 3, pp. 503–512. https://www-emerald-com.ezproxy.apiit.edu.my/insight/content/doi/10.1108/IJHMA-12-2017-0106/full/html
- Althobaiti, S., Alghumayjan , S., Frank, M., Moro, E., Alabdulkareem, A., & Pentland , A. (2021 May 31). Housing Prices and the Skills Composition of Neighbourhoods. Frontiers in Big Data. https://www.frontiersin.org/articles/10.3389/fdata.2021.652153/full
- Bello, M. Z., Sanni, M. L., and, Mohammed, J. K. (2020 December 17) CONVENTIONAL METHODS IN HOUSING MARKET ANALYSIS: A REVIEW OF LITERATURE. Baltic Journal of Real Estate Economics and Construction Management, pp. 227–241. https://doi.org/10.2478/bjreecm-2020-0016
- Delmendo, L. (2019, May 27). Pakistan’s high inflation pushing up house prices. Globalpropertyguide.com. https://www.globalpropertyguide.com/Asia/Pakistan/Price-History
- Dixon, A. (2019, July 23). How to Figure out How Much You Should Charge for Rent. SmartAsset. https://smartasset.com/mortgage/how-much-you-should-charge-for-rent#:~:text=The%20amount%20of%20rent%20you
- Globe Estate & Builders (December 2020) https://globeestatebuilders.com/pakistans-2021-real-estate-prospects-and-challenges/
- Hussain, T., Abbas, J., Wei, Z., & Nurunnabi, M. (2019). The effect of sustainable urban planning and slum disamenity on the value of neighbouring residential property: Application of the hedonic pricing model in rent price appraisal. Sustainability, 11(4), 1144. https://doi.org/10.3390/su11041144
- Imran, Zaman, U., Waqar, M., & Zaman, A. (2021). Using Machine Learning Algorithms for Housing Price Prediction: The Case of Islamabad Housing Data. Soft Computing and Machine Intelligence Journal. https://www.koreascience.or.kr/article/JAKO202122260767075.pdf
- Li, Z. (2021). Prediction of House Price Index Based on Machine Learning Methods. 2021 2nd International Conference on Computing and Data Science (CDS), 472–476. https://doi.org/10.1109/cds52072.2021.00087
- Liu, Y., He, S., Wu, F., & Webster, C. (2010). Urban villages under China’s rapid urbanization: Unregulated assets and transitional neighborhoods. Habitat International, 34(2), pp. 135–144. https://www.academia.edu/25430724/Urban_villages_under_Chinas_rapid_urbanization_Unregulated_assets_and_transitional_neighbourhoods
- Nguyen, J. (2020, October 21). Is there a correlation between inflation and house prices? Investopedia. https://www.investopedia.com/ask/answers/correlation-inflation-houses.asp
- Open Data (2020, May 15). Zameen Updated.csv. https://opendata.com.pk/dataset/property-data-for-pakistan/resource/8c882a14-ee3f-4534-a15f-5fec2319d5e5
- Phan, T. D. (2018). Housing Price Prediction Using Machine Learning Algorithms: The Case of Melbourne City, Australia. 2018 International Conference on Machine Learning and Data Engineering (ICMLDE). https://doi.org/10.1109/icmlde.2018.00017
- Sigmaproperties (2021, April 8). Pakistan Real Estate Forecast 2021. SIGMA Properties & MKT (PVT) LTD. https://sigmaproperties.com.pk/pakistan-real-estate-forecast-2021-2/
- Umar, M., Akhtar, M., & Shafiq, M. (2019). Impact of monetary policy on house prices: case of Pakistan. International Journal of Housing Markets and Analysis. 10.1108/IJHMA-12–2017–0106
- Wiebe, J. (2021, February 8). How To Calculate Property Tax: What All Homeowners Should Know. Real Estate News & Insights | Realtor.com®. https://www.realtor.com/advice/finance/how-to-calculate-property-taxes/
- Zameen.com. (2021). About Zameen.com — Pakistan Real Estate Portal. Retrieved August 6, 2021, from www.zameen.com website: https://www.zameen.com/about/aboutus.html
- Zawadzki, S. J., Mainzer, S., McLaughlin, R. A., & Luloff, A. E. (2017). Close, but not too close: Landmarks and their influence on housing values. Land Use Policy, 62, pp. 351–360. https://pure.rug.nl/ws/files/42670224/LUP_2015_120_Revision_1_V0_PROOF.pdf
Acknowledgment
The author would like to thank Abdul Ahad Lone, Sabri Bin Hilal, and Yuwadi Thein Naing who have been author group members in making this article.
