Daniel Fichana's picture

A predictive methodology for benchmarking car sales as a function of MSRP

One of the basic tenants of economics is lower prices equal more sales and higher prices equal less sales. If a company offers a “value” product for less money than the competitors they poach sales from said competitors. There is a rather interesting trend that becomes evident in auto sales when MSRP vs sales is plotted. This is written to offer a guidance instead of picking arbitrary numbers with very little basis for yearly sales.

This trend can be used to benchmark a car model vs its competitors within the same price range. There will always be outliers, but in general this trend works as a benchmark. As an example, if the MSRP of a car is 15 K and it sells 1,000 cars that is a failure. If it sells 500,000 cars, the car would be is a resounding success.

There are always qualifiers which have to be taken into consideration, these are either limited production, supply constrained, some other economic mechanism, or new model year/change in style mechanism at play.

The image with this story, shows the chart using MSRP and US sales from GoodCarBadCar.com

The black line shows the “average” expected number of cars in a given MSRP. The Y-axis (sales) is on a logarithmic scale since it illustrates the difference in sales with an appropriate resolution. A group of random cars were selected to develop the graph and includes both foreign and domestic cars. The basic rule of thumb is that when the price of the car is doubled, sales are cut by a factor of 4.

Where should the line be drawn for average, flop and success? There are two spots in the data where the price is the same but the marks are distributed. The Dart is selling 46% of the projected number of sales from the model. The Mazda 6 series is selling 44% of the projected number sales the from model. A sensitivity of 50% each way is average. If a car falls within +/- 50% of the projected number based on the model the sales would be considered is average.

How do other cars fit into this, are they below the line (not selling well) above the line (selling well) or around the line?

Five cars were chosen at random. These cars were the Prius, the Camry, the SmartforTwo, the Volt, the Model S and the ELR.

The Smartfortwo based on the price, should be selling on par with the grouping around the $15,000-$17,000 price mark, and 180,000 units per year would be acceptable. It sold a dismal 10,453 units in the US in 2014. At least in the US compared to other cars in a similar price range the Smartfortwo is not selling competitively. It is already at the rock bottom price so obtaining more sales is going to be difficult. Americans in general do not like this car and Smart should abandon the US market for the Smartfortwo.

How does the Camry do with this model? It is predicted to sell around 115,000 cars based on the model. It sold over 400,000 units in 2014. They priced the car too low and added features that are perceived as valuable. The Camry is a very competitive car at that price range. This is an outlier, but there are reasons for this aberration. The Camry has a reputation of being either the top pick or close tho the top pick in the affordable midsize car class. This is an example of a “value” product”.

The Prius was projected to sell 100,000 cars, it sold 122,000 cars in 2014. It sold slightly above the number of cars that it should sell in the US. It certainly is not as large of a hit as the Camry, but the Prius is about average in terms of sales. The Prius also has brand recognition and very little competition. The fact that it lacks competition means that it should be much closer to the higher end of the spectrum in terms of sales, but it is closer to average. This does not bode well for long term sales and survival of the car.
The EVs and PHEVs are a little dicey since the tax credits are not factored into the equation since it can be assumed that the buyers receive the tax rebate, but there is no way to determine exactly how many buyers had the tax liability and were able to take advantage of the credit. This assumes the person paid full price for the car. This has the effect of lowering the metric of sales to determine if the car is a success or failure.

The Volt should have sold close to 48,000 cars in the US last year. The data from 2014 is invalid due to a number of factors at play such as the Bolt announcement and the 2016 upgrade. The sales from 2013 would be better as a comparison. The Volt in 2013 sold close to 25,000 cars in the US. This is below the “goal” but still meets average. Contrary to popular belief the Volt is not a failure, it is just selling less than the average, but still within the margin of error for average sales.

Tesla Model S is a difficult one to figure out since most owners did not get the 60 kwhr pack. The lowest price point should not be used since the vast majority of owners purchased higher end cars. It has been reported that the average purchase price per Model S is around $100,000. At this price the Model S should sell close to 2,000 cars per year. At the $100,000 price point Tesla is selling closer to 16,000 cars per year. The Model S is selling in volume similar to cars which are 45% less. How can this be explained? This is another oddball similar to the Camry, but for different reasons. This increase in sales may be due to a variety of reasons which include various state incentives, looks, quickness, technology, convenience to fuel, and lower cost of fuel. The information that can be gleamed from the Tesla is that people want a good looking, technologically advanced, fast car with low fuel cost. It does not appear that Tesla is poaching significant sales from other high end direct competitors such as Audi, Mercedes, or Porsche. Since the Model S was introduced, the Panamera is down 1600 sales per year, the A7 has a sales decrease of 300 sales, and the the A8 is stable with sales. From the number of purchases, these purchases appear to be “stretch” purchases and poaching sales from cars that are not in the same price range as the Tesla Model S.

The ELR is a different story. Recently GM marketing came out and said they messed up on the price point. At that price point, if it was competitive with other models in that price range it should sell around 6,500 units per year. Unfortunately the ELR only sold 1310 units in 2014. Was the price point wrong or were there other reasons why the car could not compete at those levels? At $75,000 a car is expected to behave similar to an Audi A8, Equus, and to a lesser extent their own XTS.

The ELR falls into the competition trap within the Cadillac brand. A Cadillac dealership may have 10 types of Cadillacs in the showroom. This causes a problem for the ELR in two ways. The ELR is priced high so sticker shock will be a factor. Secondly, it is weak in the performance aspect compared to the other cars in the showroom. As an example, the XTS is placed in the general vicinity of an ELR, a consumer is naturally going to be drawn to the XTS over the ELR due to pricing and specifications. The customer may do a little math in their head and come to the conclusion that the premium for the ELR would not equate to gas savings over the XTS. The XTS is a stronger, less expensive car. The ELR caters to a very specific audience which is approximately 2000 customers or less per year.

Projecting future car sales

This model can be used to project future car sales and determine if the company is making a pessimistic, realistic or optimistic assumption. Back in 2011, there was a report that stated the Leaf, Volt and Focus EV would sell 100,000 cars per year at the price point of $35,000-$40,000 each. All cars can not be hits, someone has to lose. Those numbers were very optimistic and had no basis in reality. This was further compounded by the fact that the Leaf and Focus EV were rather limited, further cutting down the potential market.

Morgan Stanley projected the Model 3 to hit a production of 220,000 worldwide cars in 2020. This assumes a 4% growth per year in new cars sales and that Tesla can sell the car for 35K. When the model is adjusted for market growth and the ratio of US sales vs worldwide sales is taken into account, the Morgan Stanley number is on the pessimistic side. The Morgan Stanley number still falls within the red bounds, but when investing, it is best to choose the average case. If the Model 3 is average, worldwide, it should sell around 315,000 cars per year in 2020. If the Model S is used as a benchmark of success, the Model 3 should sell close to 575,000 units.

Subscribe to Torque News on YouTube.

Follow Torque News on YouTube, Twitter and Facebook.