Analyzing the United States Housing Market
Authors: Trent Bellinger (Project Lead), Johan Chua, Tony He, Aral Muftuoglu, Kevin Ngo, Brandon Tran
Introduction:
In this article, we will analyze many aspects of the United States Housing market. We will first explore some general trends throughout the United States housing market, specifically looking at housing supply per capita, number of bedrooms, and Airbnb listings. We will then more specifically consider California and Los Angeles housing.
How Does Housing Supply Per Capita Affect House Prices?
Recently, housing prices have increased dramatically, especially during the pandemic, raising concerns that people will not be able to afford to buy a home. There is a commonly held belief that building more houses will mitigate this through supply and demand, where a higher supply will reduce prices if demand remains consistent. My research analyzed the correlation between housing supply per capita and housing prices adjusted for inflation. It analyzed data from 1990 to 2021 and found that there is a weak negative correlation (r=-0.45) between the two. This linear relationship and the relationship between housing price index and supply per capita are shown in the visualizations below.
Note: The housing supply per capita is scaled by 200 times to make changes easier to view.
These observable trends suggest that the supply of housing may have a slight impact on prices, but there are other factors at play as well. For example, housing prices dropped significantly following the 2008 recession, which was caused by the housing bubble burst. Housing supply per capita had also increased during that time, so the recession or some other external factors may have caused both. Additionally, outliers in the data show that housing prices can remain despite a moderately high housing supply per capita. Because of this, these results do not definitively prove that more housing construction will decrease housing prices, but there is at least some negative correlation between them.
How Does the Number of Bedrooms in a Home Affect Its Price?
Another factor to consider is the average price range in terms of the number of bedrooms a home has. The boxplot below graphs the price per night for Airbnbs (to approximate rental rates) as a function of the number of bedrooms that home has.
As expected, there is an increasing positive proportional trend between the two variables, as — logically — the larger the home, the most expensive it is. However, interestingly, the trend reaches its peak at around six and eight bedrooms, where it proceeds to display a more parabolic trend. This may be explained by the fact that after a certain threshold — in this case, a home with six bedrooms — consumers become increasingly indifferent about the number of bedrooms a home has, and begin to consider alternative factors to determine its market value.
What are Factors that Contribute to an Expensive Airbnb Listing?
To identify the key factors that contribute to an expensive Airbnb listing, we analyzed data from the Inside Airbnb database using a linear regression model. Our analysis revealed the following ranking:
- Location: The location of an Airbnb is the most influential factor in determining its price. This is because the location affects the general price range of houses in the area. In addition, location is often the primary consideration for travelers when selecting a rental property.
- Number of reviews: The number of reviews a listing has is a crucial factor in its popularity and demand, which allows owners to charge higher prices.
- Total number of listings by the owner: Listings by experienced owners are perceived to be more reliable, which makes them more attractive to potential guests who are willing to pay more for a trustworthy listing.
Next, here is a graph that displays the price variations of different types of accommodations: renting out an entire home, hotel rooms, private rooms, and shared rooms.
The prices are represented on the y-axis, while the x-axis shows the types of accommodations. As we can see from the graph, renting out an entire home is the most expensive option, with prices exceeding those of hotel rooms, private rooms, and shared rooms. Hotel rooms come in second place in terms of cost, followed by private rooms and shared rooms, which are the most affordable option.
In addition to the key factors that contribute to Airbnb pricing, we have compiled a list of the top 10 neighborhoods in the Los Angeles area with the highest Airbnb prices.
This information can be helpful for investors looking to start an Airbnb business in the Los Angeles area.
So what does this all mean for Airbnb hosts and guests? For hosts, it is important to understand that the location of their listing, the number of reviews, and the number of listens all play a crucial role in determining the price they can charge for their rental. For guests, it is important to consider these factors when evaluating rental prices and selecting a listing that meets their needs.
Are Houses in California Affordable?
Affordable housing has been a long-standing issue for the residents of California for decades. Despite the state having a median household income of $80,500, the cost of housing is not proportional to the standard of living. This is particularly concerning for low to middle income families who struggle to find suitable housing in their budget.
One of the primary factors that contributes to the high cost of housing is location. Some of California’s most prominent cities, such as San Francisco, Los Angeles, and San Diego, have significantly higher median property prices and incomes than the rest of the state. The housing market in these cities reflects the high demand for housing coupled with limited supply, resulting in soaring property prices. This makes it increasingly challenging for those who cannot afford such high housing costs to find affordable housing options in these cities.
To determine the affordability trends of houses in California, we collected data from the California Association of Realtors (CAR). We used the CAR First-time Buyer Housing Affordability Index (FTB-HAI) as the metric for determining affordability, as the index represents the percentage of households that can afford to purchase an entry-level home in California. In other words, a higher affordability index means that houses tend to be more affordable in that region.
As shown by the visualization, the Central Coast, S.F. Bay Area, and Southern California regions are consistently less affordable compared to the Central Valley, Far North, and Other County regions. From the third quarter of 2021 to the third quarter of 2022, all regions became less affordable. However, because the data only covers 3 quarters, it is unclear whether this downward trend in affordability will continue in 2023. The fluctuations of the affordability index vary less from the second to third quarter of 2022 compared to the third quarter of 2021 to the second quarter of 2022. According to the data, the top two most affordable regions to buy a home in would be Central Valley and Far North with a FTB-HAI of 43 and 40 respectively.
We will now look at income levels of homeowners in California compared to the location of their house. As seen in the figure below, there is a clear correlation between income levels and the proximity to the coast. Those who earn higher incomes tend to live in coastal areas, while those with lower incomes often live inland. This pattern, as we can see below, may be due to a variety of factors, such as job opportunities, accessibility to amenities, and lifestyle preferences.
Another critical factor that affects housing prices is personal income. While it is impossible to predict individual spending habits, there is a strong correlation between household income and property prices. By examining the distribution of income and property prices, it is evident that those with higher incomes tend to invest more in their homes. The best-fit equation of y = 41794x + 45085 in the graph below further confirms this relationship, as it illustrates that most people’s house values are significantly greater than their income.
This observation highlights the importance of understanding the link between personal income and housing prices. It also suggests that those with higher incomes are better equipped to navigate the housing market, as they have more resources to invest in their homes. This may further exacerbate the problem of affordable housing, as low to middle income families struggle to compete in a market where higher-income households have an advantage:
Given the state’s high cost of living and limited affordable housing options, the government and private sector must work together to address this issue. While there is no single solution to this complex problem, efforts to increase the supply of affordable housing, incentivize developers to build more affordable units, and support low to middle income families in their search for suitable housing are crucial steps towards ensuring that all Californians have access to safe and affordable housing.
Can We Predict the Median Housing Price for Houses in Los Angeles?
We will now look at the price of houses in Los Angeles. We will attempt to create a model that will accurately depict a correlation of median house price in LA over different years. The data that I have used starts in January of 2010 and gives the median house price for every month until December of 2022. The overall trend between house price and number of months since 2022 is depicted in the plot below.
The subset of the data when houses in LA had an average time on market of over 30 days appears to be linear. However, the overall trend is clearly exponential. We want to fit a linear model to the data, so we will transform one of the variables. This transformation is shown in the below plot.
This scatterplot portrays a strong linear association throughout, so we will create a model based on this transformation. This created model was fairly accurate, and the results of its predictions are shown in the plot below, where the red points indicate a prediction that has a discrepancy of more than $50,000 and the blue points indicate a prediction that has a discrepancy of less than $50,000.
According to the data, there is a 74.36% chance that the model predicts a median house price for Los Angeles in any given month that is within $50,000 of the true value. Extending this model to predict the median house prices (in thousands of dollars) in 2023 gives the data below, where 1 is January, 2 is February, etc.
Overall, the price of houses in Los Angeles increases exponentially over time, and can be accurately predicted using an exponential regression model. If this exponential trend continues, which is expected, houses in Los Angeles will continue to grow, and the median will likely pass $1,000,000 in the near future.
Datasets:
https://www.car.org/marketdata/data
http://insideairbnb.com/get-the-data/
https://www.kaggle.com/datasets/dhirajnirne/california-housing-data
