Finding affordable housing in Paris and its surrounding region, Île-de-France, can be a daunting task. With the high demand for rental properties, prices can vary widely and change frequently. In this blog post, we’ll dive into the latest statistics on apartment and house rental prices in Paris and Île-de-France. Whether you’re a student, professional, or family, these insights can help you make informed decisions about your next rental property in the City of Lights. If you want to get insights on the rental market in the entire France, please check out this post.
Welcome to my brand new blog! This is the first post on this blog. As a hobby, I love to build web scraping bots that automatically crawl websites and download specific data to spreadsheets or databases. It’s a fun way to gather information and fuel my passion for data analysis. One of my long-term projects is analyzing housing markets in different parts of Europe, and I’m excited to share my latest findings on this blog. In this post, I’m going to focus on the metropolitan area of Paris, also known as Île-de-France, and will only consider apartments or houses for rent. Analysis of sell offers of apartments and houses in the same region I will cover in another post later.
I mainly use Python 3 to tackle challenging tasks like web scraping. For scraping, I use the scrapy library, and for post-processing and analysis, I rely on pandas, numpy, and scipy. When it comes to visualizing data, I use either the matplotlib or seaborn library. These Python libraries have proven to be incredibly useful in my data analysis work!
Part I provides an overview of the rental housing market in the entire metropolitan area Île-de-France. In Part II, we zoom in on Paris itself, exploring the rental prices in greater detail. Additionally, Part III includes some remarks about the data clean-up procedure, regression analysis, and the methodology used to estimate the average change of rental costs. And, finally, a conclusion is made in Part IV. Enjoy!
I. Île-de-France – metropolitan area of paris
The Île-de-France metropolitan area includes 24 Arrondissements, with Paris being one of them. In this section, we will study the costs of renting apartments and houses in this region of France using the most recent data available from December 2022. To calculate the mean annual change in rental costs, we will use a second dataset from December 2021 in combination with the most recent dataset.
In the first part of this section, we will examine the data describing the rental costs of apartments, and in the second part, we will discuss the same analysis for houses in the metropolitan region. Let’s start!
I.1) Apartments in Île-de-France
Let’s begin by examining the rental prices of apartments in December 2022. In Fig. 1, we see the average rental price per square meter plotted for all 24 Arrondissements of Île-de-France. As expected, the highest rental price per square meter is found in Paris itself, with an average of nearly 35.3 €/m² per month. On the other side, the lowest value is estimated for the Arrondissement called Provins, with 14.6 €/m² per month, which is also the rough average rental cost for an apartment in the entire France. Therefore, the average square meter rental price for apartments in December 2022 in Île-de-France lays between 14.6 and 35.3 €/m² per month.
Let’s compare the distribution of apartment prices per square meter in the Paris metropolitan area with the rest of France. Fig. 2 shows a skewed normal distribution of prices in Île-de-France, the region encompassing Paris, compared to the rest of the country. The median price per square meter in Île-de-France is roughly €22, with most offers around €17, making it an expensive area to live in. In contrast, the national median price for apartments is around €14, with most offers around €11, meaning that rental apartments in Île-de-France are nearly twice as expensive as the national average.
By using both the current dataset from December 2022 and a historic dataset from December 2021, we can estimate the average change in rental prices for apartments in each Arrondissement during this period. Fig. 3 displays this average change, with warmer color tones used to indicate an increase in rental prices, and cooler color tones used to indicate a decrease. We can observe that Paris itself experienced a moderate average change of rental prices, with an increase of +1.1% during the reference period. The highest increase in rental prices for apartments occurred in Évry, with roughly +7.3%. Conversely, the Arrondissement of Versailles experienced the largest decrease in rental prices in Île-de-France, with a drop of nearly -3.0% between December 2021 and December 2022.
between December 2021 and December 2022 in %
To facilitate comparison, Fig. 4 presents both the average rental price per square meter and the mean annual change for each Arrondissement in Île-de-France. As shown, the mean rental price of apartments per square meter in December 2022 is 21.3 €/m² per month for the entire region, which is 40% higher than the national average of 15 €/m² per month. We can also see that the mean annual increase of the rental costs lays somewhere around +1.7%.
Finally, in Fig. 5, we can see how the square meter rental price of apartments behaves in relation to the aerial line distance from Notre-Dame de Paris (48° 51′ 10″ N, 2° 21′ 0″ E), which is considered the central point of Paris. Some data, acquired through web scraping of several internet housing market websites, contains the exact location of the offered apartment. A detailed explanation how to estimate the aerial distance between two GPS positions is given here. The points in this diagram are the filtered and selected data points containing GPS coordinates of the particular offer. The data is clustered by the number of rooms the apartment has to offer. We can see that the highest square meter rent price is for 1- and 2-room apartments, while 3-5 room apartments cost less per square meter. We also observe that the rental price starts to increase drastically as you come closer to Notre-Dame de Paris. You need to live more than 30 km away from Notre-Dame de Paris to pay, on average, less rent per square meter for an apartment than the entire average for Île-de-France (21 €/m²/month), as seen in Fig. 5.
from Notre-Dame de Paris in December 2022
The average square meter price in relation to the distance from Notre-Dame de Paris was fitted using the curve_fit function from the scipy library package in Python 3. The underlying fitting function is as followed:
p(x) = \frac{A}{e^{B\cdot x}}+C
In a mathematical fitting procedure, the goal is to determine the values of the unknown coefficients A, B, and C, such that the deviation between the real data points and the resulting function is minimized on average for all data points within the given subset. The resulting coefficients are shown in the following table:
| Subset | A | B | C |
|---|---|---|---|
| All data points | 26.502 | 0.082 | 18.390 |
| 1-room data points | 25.409 | 0.057 | 22.229 |
| 4-7 rooms data points | 18.575 | 0.079 | 14.924 |
in relation to the distance from Notre-Dame de Paris
Coefficient C is a minimum limit price that the functions approach but never exceed as the distance from Notre-Dame increases. Tab. 1 shows that for 1-room apartments, this limit is 22 €/m² and for 4-7 rooms apartments, it is lower at 15 €/m². The average C coefficient for all data is around 18 €/m². Generally, renting a 4-7 room apartment is cheaper per square meter than a 1-room apartment because not all costs scale uniformly with the size of the apartment. Fixed costs such as council taxes are included in the monthly rent, resulting in higher rent per square meter for smaller apartments.
I.2) Houses in Île-de-France
I performed a similar analysis for rental houses in Île-de-France. Therefore, in Fig. 6 the average rental price per square meter across the 24 Arrondissements is shown. Some Arrondissements are greyed out due to insufficient data required to establish a reliable average rental price per square meter. The dataset had fewer house rental offers compared to apartments. Thus, a minimum of 10 data points per Arrondissement was deemed necessary for a reliable average square meter rental price for the particular Arrondissement. Please note that the reference area for the rental price is the indoor living area of the house, not the exterior total ground area.
By comparing to Fig. 1, we can see that the average rental price per square meter of houses is less compared to apartments. Inside Paris it is very rare that a house if offered for rent. Moreover, Nanterre and Boulogne-Billancourt are the Arrondissements with the highest rental prices of houses with more than 23 €/m²/month. On the other side, the same as for the apartments, the cheapest Arrondissements to rent a house is Provins with 11 €/m²/month.
Like Fig. 2, Fig. 7 shows the distribution of house prices per square meter in Île-de-France and the rest of France. The median price in Île-de-France is lower than for apartments. However, renting an entire home in this region costs nearly €18/m², compared to the national average of roughly €10/m². This means that renting a whole home in Île-de-France is 80% more expensive than the national average.
The annual growth of rental prices is presented in Fig. 8. The data shows that rental prices in Torcy, Anony, and Évry have increased by more than 9% between December 2021 and December 2022, whereas rental costs of houses in Nogent-sur-Marne, Étampes, and Saint-Denis have dropped by more than -5% during the same period.
December 2021 and December 2022 in %
Therefore, in Fig. 9, one can see the actual rental price of houses per square meter and the corresponding mean annual change for every Arrondissement. We can see that per square meter, it is definitely cheaper to rent a house than an apartment, as the mean rental price per square meter is roughly around 17.6 €/m²/month, while for apartments it’s average is around 21 €/m²/month in the entire metropolitan region (see. Fig. 3). Further, the mean annual change of the rental prices of houses in Île-de-France has increased by 2.1% between December 2021 and December 2022.
II. Paris – The City center
Now, let’s take a closer look at Paris. In Fig. 10, you can see the average rental prices per square meter for apartments as of December 2022. Unfortunately, there isn’t enough data to provide statistics for houses as they’re rare inside Paris. Additionally, there wasn’t enough data for the 2nd Arrondissement, so that information is not included. The 1st Arrondissement is the most expensive area to live in Paris, with apartments costing around 45.8 €/m² per month on average. On the other hand, the 19th Arrondissement is the cheapest area, with apartments costing nearly 31 €/m² per month. Overall, the cost of renting an apartment in Paris ranges between 31 and 46 €/m², which is 2-3 times higher than the national average of 15 €/m² per month.
Fig. 11 provides information on the average size of apartments across different Arrondissements. One interesting finding is that the smallest apartments are located in the 1st Arrondissement, with an average size of only 34 m². On the other hand, as one moves towards the outskirts of Paris, the average size of apartments increases to around 45 m².
Similarly to Figs. 3 and 8, we can also examine the mean annual change of apartment rental costs. The findings are presented in Fig. 12, which focuses specifically on Paris. On average, apartment rental costs have increased by approximately 2% annually. The most significant increase was observed in the 3rd Arrondissement by nearly +7%, while the most substantial drop was seen in the 4th and 9th Arrondissements, with a decrease of more than -3% between December 2021 and December 2022.
between December 2021 and December 2022 in %
Fig. 13 provides a detailed overview of the average rental prices per square meter and the approximate annual change for each Arrondissement. The figure shows that the average rental price in Paris is around 37 €/m² per month. Furthermore, the rental prices experienced a moderate increase of around +2% between December 2021 and December 2022.
Let’s take a closer look at the rental prices in Paris. Fig. 14 shows two plots. On the left side, the total monthly rent price is plotted against the total size of the apartment, while on the right side, the rental price per square meter is plotted against the total size of the apartment. The dots represent real apartment offers collected in December 2022. The data is clustered by the number of rooms offered in each apartment, with the majority of apartments having 1-3 rooms. It is clear that the total price is mainly influenced by the size of the apartment, although other features of the apartment also affect the price. To help determine a rough reference value for the cost of an apartment of a certain size in Paris, a linear regression analysis was performed and the resulting formula can be seen on the left plot. This reference value can be used as a guide, but it’s important to remember that a higher price could be justified by beneficial features of the apartment, such as a renovated space, a quiet environment, or a better energy rating.
Moving to the right plot in Fig. 14, we can see that 1-room apartments are the most expensive to rent per square meter, whereas apartments with 3 or more rooms achieve an optimal balance between total rent price and rental price per square meter.
For those who are planning to move to Paris for a short-term period, it can be advantageous to rent a furnished apartment. In Fig. 15, the dataset was separated by whether the apartment is already furnished or not. Furnished apartments are marked with orange dots, while non-furnished apartments are marked with blue dots. By conducting a linear regression analysis on both groups separately, we can see that furnished apartments have a slightly higher average cost. Furthermore, it is noticeable that mostly small- to mid-sized apartments are offered furnished, while larger apartments are usually offered without any furniture.
III. Remarks on Data processing
In this section, I will provide an overview of the data cleaning and post-processing techniques that were employed to refine the raw data gathered from the web. In the first sub-section, I will explain the methodology used to eliminate invalid data points from the dataset. In the second sub-section, I will describe how the mean annual growth of rental prices was estimated, as discussed in the preceding parts of this post.
III.1) Selection of valid subset of data points
Working with a large scattered cloud of data points can be challenging, especially when outliers are present. These are data points that do not fit into the general trend of the data. An example subset is displayed in Fig. 16 for Paris itself, where we can see that the total monthly rental cost is nearly linearly depended on the total apartment size. However, we can also see that some data points are located to far outside of the general trend and, therefore, are suspicious to be invalid data. To clean up the dataset and filter out non-valid data, I performed a linear regression analysis for each Arrondissement and removed all points whose rental price deviated from the reference price (estimated by the regression analysis) by more than 50%. This helped to obtain a more accurate representation of the rental market within the metropolitan area.
III.2) Estimation of Annual Change of Rent prices
As we already know, we can assume that the total rental price of an apartment or a house is depending mostly by the inhabited area. Therefore, we can state that the average price, p(s), can be described approximately as a linear function as followed:
p(s)=a+b\cdot s
where a and b are unknown coefficients and have to be estimated using standard linear regression analysis. We can perform such a regression analysis for the the data set from December 2021 and December 2022 independently for a specific subset (i.e. house or apartment in a specific area) and we will obtain p_{new}(s) and p_{old}(s). The mean average change of all apartments or houses in an area can be then estimated as followed:
\bar{c}_{p(s)} = \frac{1}{s_{max}-s_{min}}\int_{s_{min}}^{s_{max}}\frac{p(s)_{new}-p(s)_{old}}{p(s)_{old}}ds
This is an integral formulation of the average annual change over the inhabited apartment or house size s. Here s_{min} and s_{max} are the integration boundaries where 99% of the data points of the particular subset are inside regarding the total size of the apartment or a house. They can be estimated for houses or apartments and for the particular Arrondissements using statistical methods. Finally, the average approximate change of houses or apartments in a particular area can be stated as followed:
\bar{c}_{p(s)} = \frac{\left (s_{max} - s_{min} \right )\cdot\left(\frac{b_{new}}{b_{old}}-1 \right ) + \frac{a_{new}\cdot b_{old} - a_{old}\cdot b_{new} }{b^2_{old}}\log{\left(\frac{a_{old}+b_{old}\cdot s_{max}}{a_{old}+b_{old}\cdot s_{min}}\right )}}{s_{max} - s_{min}}
IV. Conclusion
This analysis provides valuable insights into the rental market in the Paris metropolitan area. It shows that rental prices have continued to increase, making it difficult for many individuals to find affordable housing. By identifying the factors that influence rental prices, this study provides useful information for those looking to rent properties in the area.
As of 2023, the cost of living in Paris remains high. On average, rental prices for houses and apartments in the metropolitan area have increased by approximately 2% between December 2021 and December 2022. The Île-de-France region continues to be the most expensive region in France to live in, with average rental prices of €21/m² for an apartment and €18/m² for a house. Within Paris itself, apartment rental prices can reach as high as €46/m² directly in the city center. In contrast, rental prices in Provins remain moderate, around €12/m².
Predicting market trends can be challenging as it can be unpredictable and change direction suddenly. However, I will keep monitoring and collecting data on housing market offerings from websites and provide updates on future housing market trends on this blog. If you find this analysis helpful, please show your support by leaving a like and comment below. Please, don’t hesitate to leave comments if you have noticed some mistakes or you have suggestions on improving the content. 🙂
.jpg){kind=link}
Congratulations, Daniel, for setting up yourr own blog and this first interesting post. I’m not planning to go to Paris, but perhaps your concept can be adapted to our region as well. It would be interesting to know how the different levels of salary paid in three different countries influence on the living cost… Cheers!