top of page
  • Writer's pictureDiego Leonardis

Price determinants in the automotive market: analysis of the 'supercar' segment in the United States

The objective of this paper is to investigate the supercar segment of the U.S. automotive market and, primarily through the construction of a multiple linear regression model, highlight the elements having a statistically significant impact in defining sales value. Among the many explanatory variables encapsulated by the model, it was decided to focus on five research elements:

· State’s wealth: in areas with higher concentrations of wealthy individuals, do advertised prices increase?

· Brand equity: are there supercar manufacturers with brand equities having an impact on advertised prices?

· Color: Does the car's exterior and/or interior color impact its depreciation process?

· Certified pre-owned: are certified pre-owned vehicles on sale for higher prices?

· Seller’s reputation: do dealers with better reviews charge higher prices?

It’s possible to find evidence in literature on the statistically significant effect that the brand, color, certified pre-owned, and seller reputation variables have on the resale price of a car, in different markets and for different car segments (Cho, 2005; Andrews & Benzing, 2006; Ramachandran & Gosain, 2007; Powers, 2016; Kantar & Bardakci, 2017).


In order to pick the models to be included in the analysis, the concept of “supercar” had to be formalized: after tracing the history of this phenomenon that started with the Lamborghini Miura (unanimously considered as the first supercar in history) and analyzing opinions of journalists and experts, five key elements were identified that determined the membership of a car to the supercar segment:

· Top-of-the-range performance;

· Limited production;

· Sporty design;

· Exclusive price;

· Exciting driving experience.

Filtering the cars sold in the United States from 2014 to 2022 according to the above criteria revealed 26 models (belonging to 8 brands) to be considered as supercars.


The dataset used for the analysis was compiled through a web scraping process (meaning an automatic extraction of data from a website), in this specific case through a Python script leveraging the Beautiful Soup package.

The reference website was chosen to be cars.com, after evaluating several marketplaces by the number of car ads posted and the detail of information in each ad. The Python script was written exclusively for the purpose of this work, as there were no similar ones in terms of features and applicability to the selected website.

Running the script for all models under analysis yields an Excel file containing an initial raw dataset. After some data cleaning (e.g., elimination of outliers) and manual filling in of missing pieces of information, the final database shows 3,487 cars. For effective model building, it was necessary to estimate the original price (including optional extras) paid by the first buyer of each car.

It was decided to estimate an "options coefficient" for each car model under analysis such that, when multiplied by the list price, the total original cost of the car could be estimated with a good degree of accuracy. For example, analyzing Lamborghini Huracán ads showing the full price breakdown, on average a buyer of the model spends 29% of the base price on optional equipment: multiplying the base list price of each Lamborghini Huracán for sale by 1.29 therefore estimates the total price originally paid.


An example of the data extraction process can be seen in the picture above: starting from an ad on cars.com, the script automatically extracts the information contained in the white cells, while the gray cells are the result of the manual input described above.



The multiple linear regression model has as dependent variable the percentage price fluctuation between the estimated original total cost and the advertised price. The independent variables are all the characteristics of the cars, sellers and market extracted through the process outlined above.






Analyzing the coefficients concerning the research questions, a higher percentage of millionaire households has a slightly negative impact on the advertised prices of U.S. supercars, probably due to the larger number of such cars available and sold in these territories.

Regarding color, black exteriors and blue-colored interiors have an accentuating effect of the depreciation process, while a green exterior color and white interiors appear as mitigating factors in the supercars’ loss of value.

Non certified pre-owned cars are subject to slightly stronger depreciation.

McLaren, Aston Martin, and Maserati branded cars experience an accelerated depreciation process, while in contrast the Porsche brand appears to have a positive influence on price fluctuation.


Through uni- and bi-variate analyses and the multiple regression model, three managerial insights can be proposed:

· Color trends: the results of the analysis appear to support the idea that color trends in the luxury sector (i.e., green, at the time of writing) have a statistically significant effect on the price of cars offered to the public. Assuming that this positive delta is then confirmed during the actual purchase, buyers are willing to spend a greater amount of money to own a car in a fashionable color. Analyzing qualitatively the promotional strategies of the brands in scope, no extra attention seems to be paid to color trends in the luxury world: this could be a relevant element to consider in future marketing and communication campaigns.

· Localized marketing: the data clearly shows the great weight (over 35%) that the states of Florida and California have on the market of supercars for sale in the US. This potential appears, at the moment, not to be fully exploited by brands: analyzing the marketing campaigns of the manufacturers involved, no particular attention to these two states stands out.

· Brand equity: as described, Aston Martin, McLaren and Maserati have substantial shortcomings regarding this aspect. Although all brands are surely prestigious, the reasons for poor performance in this area can be identified in low reliability for McLaren and not up-to-date technology for the other two brands. The manufacturers' strategies appear to acknowledge these shortcomings and work towards solving them: Aston Martin and Maserati, with the new Valkyrie models and the revamped GranTurismo range respectively, place great emphasis on the technology present on board, while McLaren offers a three-year warranty on all new models sold.


In conclusion, the model remains open to improvements and refinements, for example, developing methodologies for a more effective estimation of the original total price, or integrating databases reporting the final price agreed upon, and actually paid, by the buyer.




22 views0 comments

Comentarios


bottom of page