Do Halo Cars Change Consumer Behavior Enough to Justify Their Development?

"Tim (Fractal Footwork)" (fractalfootwork) 05/06/2014 at 10:00 • Filed to: Economics, Halo Cars, OppositeLock

39

75

This was a research paper written for my capstone in Economics, and remains the only paper I've ever been proud of. It might be a little thick, but I thought I'd post it seeing its pertinence.

Introduction

Automotive manufacturers sometimes choose to introduce a halo car, a car that signifies the pinnacle of its current technology. However, automotive manufacturers often choose to either abandon their halo model or not to develop one entirely, particularly in light of their concern for the halo car's relevance in the brand's overall market. The potential effects of a halo car in providing more efficient capital investment and in modifying the consumption of lesser models, through both brand perception and trickle-down engineering, to determine how they affect developing firms' market share within an oligopolistic automotive market will be argued here by measuring those firms' market share in relation to their exploitation of halo models.

Literature Review

Four research articles are relevant as respective associations to my research topic; one defines a "brand halo" on the used car market, another shows brand-image associations in the automotive market, the third notes preference signaling of consumer choice in the environmentally friendly automotive market, and the last measures sequential capital investment in the competitive environment of a farm.

In "The 'Brand Halo' Effect: Brand Reliability Influence on Used Car Prices," Betts and Taran (2002) discuss the effects of a brand halo on the used car market, finding a positive correlation between brand reliability and the price of a subset vehicle of that brand. The researchers regression supported their hypotheses that "wheelbase, engine size, price of a new model moderated by age, reliability, reliability moderated by age, and brand reliability are each significantly related to the used car price," using brand reliability as the brand halo variable and the other six as controls (22). They also found a correlation between reliability and brand reliability, suggesting that the effect of brand reliability on the price of a used car "was beyond the effect of the specific reliability of the individual model of used car" (22). This research is important to the topic because it shows the "halo effect" taking place and changing the consumption of certain vehicles in the used car market relative to what the car's actual reliability would have done to determine its level of consumption; this suggests that consumers are paying more for a falsely perceived idea of reliability (22). I plan to focus my research around halo effects other than reliability, but Betts and Taran's research suggests that consumers do take into consideration the perception of the brand as a whole when consuming automobiles in the competitive environment of the used car market.

In "Estimating the Value of Brand-Image Associations: The Role of General and Specific Brand Image," Sonnier and Ainslie (2011) attempt to "demonstrate the substantial value of specific brand-image associations" (519). Sonnier and Ainslie note the two brand differences in automotive models, one for the parent company and another for the model name, used to separate brand imagery between the two names; the parent brand is encompassing and the model name sounds out the attributes of that model (520). The researchers compare the "willingness to pay" for a "two-factor model" of excitement and dependability and find a negative correlation for certain models, and no correlation for others (519, 525). The research points out the separation of brand-imagery between two axes of data, and this is important because it exemplifies the fact that a consumer is willing to sacrifice one factor for another; this may be hampering to my results if I choose to regress solely one 'halo' factor, such as performance, and disregard another, such as environmental promotion. One brand may produce a new 'halo' performance car, but the sales data for the rest of the brand's range may not move due to a shift in consumer preference toward another factor.

Sexton and Sexton (2012) in "Conspicuous Conservation: The Prius Halo and Willingness to Pay for Environmental Bona Fides" note the Toyota Prius's involvement in "conspicuous conservation," a consumer's desire to "undertake costly actions in order to signal their preferences for the environment" (Sexton 2). The researchers "test empirically for the presence of a conspicuous conservation eect in vehicle purchase decisions" in communities that hold an environmentally friendly ethos, determining green communities as those who vote Democratic (3). They also note a "green halo," the willingness to pay for an obviously green car over a covertly green car, proposing "green signaling [as] an alternative" to "conventional altruism" before testing it with empirical data (6, 10). The difference between overt, obvious halo models and covert models is the immediacy in recognition of a car's qualities; for example, the Toyota Prius has recognizable and individual bodywork and is therefore overt, whereas the Toyota Camry Hybrid has bodywork that looks like any other Toyota Camry with a standard gasoline powertrain. If people are willing to pay for an image of a hybrid vehicle over the actual carbon emissions saved, then consumption is aided by brand image in certain markets and a halo effect of that brand image may be able to steer consumer behavior. The researchers then found "that Prius is likely being substituted in place of the Civic Hybrid and Camry Hybrid" due to their absence of a "unique design" and "the value of the green halo signaled uniquely by the Prius" (15). Because the Toyota Prius, and other current green halo models, are closer in price to the manufacturer's base models than performance models, something besides the halo's advertisement of the brand may have to influence a consumer's decision to purchase a lesser model.

Where Betts and Taran (2002) and Sonnier and Ainslie (2011) both enter their studies with obvious distinction for the brand and the subset brand of vehicles, Sexton and Sexton (2012) begin their research with the thought that less obvious hybrid vehicles will be consumed normally in green communities, only to later gather that the covert hybrid vehicles under the same brand as the overt hybrid vehicle are sold to a lesser degree in green communities. Betts and Taran (2002) measure their data in the competitive environment of the used car market, which should make their empirical statistics more accurate in testing their hypotheses, but Sexton and Sexton (2012) measure their data in the new car market where supply side constraints could limit companies from producing to the same degree as another, potentially yielding their data inconclusive.

Of the three articles cited, one that defines a "brand halo" on the used car market, another that shows brand-image associations in the automotive market, and a third that notes preference signaling of consumer choice in the environmentally friendly automotive market, the latter two will help direct my theoretical analysis while the former two will provide a basis for testing my hypothesis that halo cars change consumption of lesser models.

Theory

The potential for a firm's increase in demand due to the introduction of a halo model lies with the halo model's potential to change that firm's brand perception through a "brand halo" (Betts 20). When a "brand halo" exists, a consumer evaluates "a good based on the level of some attribute generally associated with the brand," disregarding in part "the actual level of the attribute present in the specific model being evaluated" (20). If a halo model is introduced into the market, is able to change brand perception, and all other variables are held constant, then a shift in demand should follow that change in brand perception.

In an Oligopoly market, where only a limited number of firms sell products, larger firms usually have the power to control the output of smaller firms based on how the larger firms set their own output. In this type of market, an individual firm's best chance to increase profit is to steal market share away from another competing firm; a way to accomplish this without having to respond to another firm's direct price and output would be to increase demand through advertising or decrease marginal cost and have the other firms respond to your strategy. A decrease in marginal cost can lower the price of a product, an increase in demand can increase the price, and a combination could do either; however, both changes can increase quantity. In both cases, consumer surplus increases, the area under the demand curve minus the area under the price curve, and producer surplus increases, the area under the price curve minus the area under the marginal cost curve. The consumer surplus measures the cumulative benefit the consumers who purchased a product gain from buying that product, and the producer surplus measures the benefit the producer gains from selling; these are important to show as a result of a halo car's induction because they display how much benefit is added on both sides without physically changing a base automobile.

The learning curve illustrates that "costs fall with cumulative output" in "a wide range of industries, and management consultants have stressed the importance of learning for production planning" (Majd 331). Because "firms face considerable uncertainty about future demand," risk rises when implementing the learning curve on one factor, as it may turn out not to reflect future demand (Majd 332). For these reasons, it may be more effective for automotive firms to test future development on the smaller market of a halo model. By initially investing in capital to produce a halo model, firms are able to both accumulate physical, human, and reputational capital and gauge how effective that capital is without overinvesting in inefficient capital when the firm refreshes the base models. For example, if a firm predicts that the future consumer may demand an electric drivetrain from a base model, the firm can first purchase and test equipment used to make the electric power units on a halo model and then purchase more capital if demand for electric drivetrains increases in the average model. This way, the firm's knowledge of building electric cars along with their expertise in purchasing efficient equipment reduces the total cost in the second period compared to if the firm did not produce a halo model in the first period.

Much in the same way that an advertisement might change a firm's perceived quality, a halo car may also have the power to modify consumer behavior, but a halo car's potential to change demand for a firm's range of cars may vary based upon how the halo model is presented to the consumer. According to Settle and Golden's (1974) take on Attribution Theory, "If the message is attributed to the advertiser's desire to sell, the consumer would be uncertain about the actual characteristics of the brand and the probability of her purchasing it would be expected to decrease," while "An attribution to the actual characteristics . . . would be expected to lead to a higher certainty and a higher probability of purchase of the brand" (Settle 181). In other words, the probability that a consumer would purchase a base automotive model prior to learning about that brand's halo model hinges on the consumer's perception of shared qualities between the halo model and the base model and the absence of the halo model's perceived existence as purely an advertisement for the brand (Settle & Golden). A halo model may also benefit from exclusivity over that of a celebrity endorsement. Where a celebrity could be offered an exclusivity agreement they may not be willing to sign, who's "nondistinctiveness [by endorsing multiple brands] may result in consumers' inferring that the nature of the spokesperson was the reason for the endorsement, not the nature of the product", a halo model's exclusive design is in the hands of the brand (Tripp 536).

Empirical Analysis

While there are more than a handful of candidate brand halos acting on an automotive manufacturer, I will specifically be measuring two, as they are highly apparent and therefore more easily measurable. The two candidate halo models used will be considered a halo over the brand if it is an original model and it exemplifies a pinnacle of performance or environmental status far beyond that of the brand's current offerings. The halo models will be broken down into two categories, green and performance, where green models will overtly display environmental conservation and performance models will overtly display a pinnacle of speed. The market share data comes from WardsAuto.com and spans the time period from 1961 to 2013 for brands selling within the United States.

I use a time-series regression model with lagged dependent variables to analyze the effect of a halo model on market share. The lag of the time period exists to see if a halo model has a lasting effect on market share after the halo model has finished production. The lag will extend back 6 years. An automaker's market share, denoted as MS , is the dependent variable, while the independent variables for this model are the existence of an environmentally conscious halo model, denoted as GreenHalo , the existence of a performance halo model, denoted as SpeedHalo , and the number of firms in the market selling more than .01 percent of vehicles, denoted as Firms . The existence of a halo model will be run as a dummy variable with values of only 1, if the halo car exists, or 0, if the halo car does not exist. The time-series regression model will be run as follows.

MS t =  0 +  0 GreenHalo t +  1 GreenHalo t-1 + ... +  n GreenHalo t-n +  0 SpeedHalo t +  1 SpeedHalo t-1 + ... +  n SpeedHalo t-n + Firms t

The hypothesis I will be examining will measure if an automotive manufacturer’s market share changes depending on whether or not that manufacturer is producing a halo model. The empirical model also takes into consideration how a halo model may affect a change in market share over time, even after the halo model is no longer in production, to check for a possible decay on market share.

The standard hypothesis states that the existence of a halo model is not able to explain a change in a manufacturer’s market share.

H 0 :  n ,  n = 0

The alternative hypothesis states that the existence of a halo car may explain a positive change in a manufacturer’s market share. I also expect the effect to converge to 0 the greater the lag, the time since the halo model finished production.

H a :  n ,  n > 0

After running the regression model, as seen on Table 1, with both the green halo and performance halo models as independent variables with a lag of 6, as that was the maximum the program would allow, the performance halo model seems to have an immediate effect on market share and then diminish with time. However, the confidence of that effect immediately decreased with the lag, the green halo model looked more sporadic than the performance model, and the last lag was much higher than I expected on both the performance and the green models, so I ran a second set of regressions with a greater lag and with the performance and green halo models separated.

Table 1

The second set of regressions, as seen on Table 2 for green halo models and Table 3 for performance halo models, continued to produce sporadic coefficients for green halo models and a low confidence level for both performance and green halo models’ effect on market share. However, the final lag seemed to spread its effect over the newly added preceding lag variables. The data proves to be relatively inconclusive for both green halo models and performance halo models affecting market share, but does reflect positively on performance models’ effect as the coefficient for the performance halo is positive with a p-value less than .05.

Table 2

Table 3

I hypothesized that the halo model would have the biggest effect in the time of measurement and decrease thereafter; this does not turn out to be accurate for either halo model from what the regressions have shown, and may be a result of the decrease in marginal cost more than a change in demand. The green halo models’ existence does not seem to show any concrete positive or negative correlation with market share. However, the performance halo models do show a positive correlation with market share in the current time period, but does not show any significant correlation thereafter; this could be explained as an even distribution of exposure of the halo model to the consumer over a number of years, or a linear build up of reputation.

These results fall in line with most of the previous research, when considering performance halo models. Although, when recalling green halo models, these results measured the entire United States market while Sexton & Sexton’s (2012) research factored out environmentally supportive Democratic buyers from Republican buyers that showed a “disproportionate increase in Prius market in green communities” (15). Where “conspicuous conservation” is apparent for democratic buyers to signal their approval of environmental conservation, the opposite could also be true for republican buyers trying to signal their disinterest in environmental vehicles, potentially explaining the difference in our results (Sexton & Sexton).

Conclusion

In quest to increase market share, some halo cars may have the potential to change demand as a form of advertising and change total cost through more efficient capital investment in an oligopoly market. These empirical results suggest that a performance halo model is more able to increase market share than environmentally conscious halo cars in a region of evenly mixed demand. These results suggest that firms do not currently have an incentive to produce environmentally conscious halo cars in a politically even region in an attempt to increase overall market share, but may have the incentive to produce overt green cars to steal sales away from covert green cars, which is consistent with Sexton & Sexton’s (2012) research.

If a government wishes to protect the environment for paternal reasons and a country’s indifference toward environmentally conservative cars by its citizens exists, then that government may have to enact regulations to produce the desired result. However, countries like the United States have already enacted these sorts of environmentally protective standards, “which mandate an average fuel [consumption] of 54.5 miles per gallon for the 2025 model year,” on automotive models, and these established laws affecting all cars may currently be influencing consumers’ indifference toward overtly green halo cars (Vlasic).

This research could be expanded to measure not just a performance or green halo model’s effects, but to measure consumers’ level of a certain characteristic on a brand and how that affects the price of a series of models. For instance, before 2014, BMW had not produced an original performance halo model since the 1980s, but they do produce performance-modified versions of their regular models; this could potentially create a performance brand halo for BMW that should be measured. Other questions related to this topic that deserve attention include how other firms respond to the introduction of a halo model by a competing firm, how market entrants are affected by introducing a halo car, how a halo model affects the elasticity of demand for regular models, how smaller niche supercar companies owned by larger firms affect the larger firms’ change in demand and total cost, and how a combination of two or more halo characteristics in a halo model affect market share.

The effects of a “brand halo,” as noted by Betts & Taran (2002), have the ability to modify an equilibrium of a base product though changes in how the consumer perceives the brand selling that base product. These changes have forced some large firms to produce halo cars in certain periods but not in others, and understanding and being able to predict why may allow other firms to lead or react to another firm introducing a halo car, adding even more depth to simple output and price changes.

References

Betts, S. C., & Taran, Z. (2002, April 13 ). The 'brand halo' effect: Brand reliability influence on used car prices. Proceedings of the Academy of Marketing Studies , 7(1), 19-24. Retrieved from !!!error: Indecipherable SUB-paragraph formatting!!!

Majd, S., & Pindyck R. (1989). The learning curve and optimal production under uncertainty. RAND Journal of Economics , 20 (3), 331-343.

Settle, R., & Golden, L. (1974, May). Attribution theory and advertiser credibility. Journal of Marketing Research , 11 (2), 181-185.

Sexton, S. E., & Sexton, A. L. (2012, October 5). Conspicuous conservation: The prius halo and willingness to pay for environmental bona fides. Journal of Environmental Economics and Management , 67(2). Retrieved from !!!error: Indecipherable SUB-paragraph formatting!!!

Sonnier, G., & Ainslie, A. (2011). Estimating the value of brand-image associations: The role of general and specific brand image . Journal of Marketing Research , 3(48), 518-531. Retrieved from !!!error: Indecipherable SUB-paragraph formatting!!!

Tripp, C., Jensen, T., & Carlson, L. (1994). The effects of multiple product endorsements by celebrities on consumers' attitudes and intentions. Journal of Consumer Research , 20, 535-547.

Vlasic, B. (2012, August 28). U.S. sets higher fuel efficiency standards. New York Times . Retrieved from !!!error: Indecipherable SUB-paragraph formatting!!!

(2014). U.S. vehicle sales market share by company, 1961-2013. WardsAuto . Retrieved form !!!error: Indecipherable SUB-paragraph formatting!!! .

Kailand09 > Tim (Fractal Footwork) 05/05/2014 at 21:25

0

I'll be honest, I haven't the time to read this right now. But wow. From what I've read great work, I will certainly take a closer look later on. Sweet stuff.

OPPOsaurus WRX > Tim (Fractal Footwork) 05/05/2014 at 21:42

0

i'm supposed to be working so I can't read this either but my opinion, their existence justifies them self. report this at lunch tomorrow (eastern time)

Somethingwittyer likes noisy > Tim (Fractal Footwork) 05/05/2014 at 21:44

10

Damn it Fractal why must you shame us Oppo users with your fantastic OC. All I do is post stuff I find on Youtube and here you are just writing properly sited economic reports and giving some of the best F1 Tech coverage on the internet. You've shamed to the point where I feel I need to finally write a legit article on Formula Freak. Damn you!

Tim (Fractal Footwork) > Somethingwittyer likes noisy 05/05/2014 at 21:47

0

Haha, I forgot you are an author.

samdonnell > Tim (Fractal Footwork) 05/05/2014 at 21:48

1

Is there a Tl;dr?
juuuuuust kidding. from what i read, looks like you wrote something big and important. Keep it up!
Wish i could focus long enough to accomplish something like this.

Tim (Fractal Footwork) > samdonnell 05/05/2014 at 21:49

2

tl:dr is basically yes, but only for performance halo cars, not so much for green halo cars.

Somethingwittyer likes noisy > Tim (Fractal Footwork) 05/05/2014 at 22:08

0

That long ass FD article finally paid off :D

quarterlifecrisis > Tim (Fractal Footwork) 05/06/2014 at 10:08

0

Was this the one we were discussing on here a few months back?

getchapopcorn > Tim (Fractal Footwork) 05/06/2014 at 10:32

4

What software are you using that limited your function's variables so highly? Surely not STATA. Also, my nose seems to be detecting the smelly smell of heteroscedasticity.

Tim (Fractal Footwork) > quarterlifecrisis 05/06/2014 at 10:45

0

Yes

Tim (Fractal Footwork) > getchapopcorn 05/06/2014 at 10:46

0

Excel, only allows 16 x variables I think. I probably should have used SAS, but oh well.

KnowsAboutCars > Tim (Fractal Footwork) 05/06/2014 at 10:48

0

I was going to read this from start to finish (honestly) but then I saw those formulas and decided to give up.

quarterlifecrisis > Tim (Fractal Footwork) 05/06/2014 at 10:59

1

Well done. I hated econ and stats....and enjoyed the read. Likely attributed to interest in the subject. Honestly, it's been at least 4 years since I've stared at tables like that...yet it still comes back. Sort of.

Honesty > Tim (Fractal Footwork) 05/06/2014 at 17:14

0

Make enough of them and they do.

Doctor-G-and-the-wagen > Tim (Fractal Footwork) 05/06/2014 at 17:14

0

Well we CLEARLY know by now they won't make ladies attracted to you.

Victorious Secret > Tim (Fractal Footwork) 05/06/2014 at 17:20

1

Masterchief approves of this thread

DMCVegas > Tim (Fractal Footwork) 05/06/2014 at 17:21

1

tl;dr. Condensed version:

Win on Sunday, Sell on Monday.

Desu-San-Desu > Tim (Fractal Footwork) 05/06/2014 at 17:21

6

Yes. You have no idea how many times I've waited in line for hours and hours well into the midnight hour in order to be the first on my block to lay down my hard-earned cash on a chance to drive a Halo car.

Diesel > Tim (Fractal Footwork) 05/06/2014 at 17:22

0

So, what you're saying is the XL1 should be cancelled before they start.

Diesel > Doctor-G-and-the-wagen 05/06/2014 at 17:24

0

My car is a strip club.

Dollar dollar bills ya'll.

Formula1_Chris > Tim (Fractal Footwork) 05/06/2014 at 17:25

0

Hell yes they do. Look at the Audi R8 when it pretty much debuted in the first Iron Man movie. It had LED's and an aggression to it and since 2007 Audi allocated the LED's to every car in their line-up. Boom! Record sales. This is because people thing, hmmm not only do I have the same badge as that awesome R8 but I have the same kind of headlights too. And I only paid 40 thousand for mine. I don't have an Audi btw

justregisteralready > Tim (Fractal Footwork) 05/06/2014 at 17:26

0

*Starts reading*

*Scrolls down*

Hmm...

*stops reading*

*scrolls....scrolls.....scrolls....*

What the..

*Scrolls..*

*Sees algorithms...scrolls*

*Sees charts...scrolls*

More words!

*Scrolls*

And REFERENCES?!

Holy crap, dude! I'm not reading all that! But kudos to you for putting that much effort into it. I'm sure it was a good read.

Thud Slamrod > Tim (Fractal Footwork) 05/06/2014 at 17:26

0

Most people can't name a Mitsubishi model that isn't the Evo, but Honda's been fine without the NSX for 10 years. If your brand lineup consists of drying paint cinema, no halo car is going to help.

Viljo Saari > Tim (Fractal Footwork) 05/06/2014 at 17:28

1

As a fellow car nut who recently switched to an Economics major, this is awesome.

Just Cars for Joe Bryant > Tim (Fractal Footwork) 05/06/2014 at 17:31

1

Hmmmmm, halo cars. Well almost 62 years and counting baby!!!!

PatBateman > Tim (Fractal Footwork) 05/06/2014 at 17:31

0

I read it. Fantastic work.

I believe that, in the next 10 years, the covert green and performance halo cars will start to merge into one category. We're starting to see that now.

How did you come up with the formulas?

Humpty-D > Thud Slamrod 05/06/2014 at 17:35

0

Honda has been struggling for a decade or so now, though one could argue that it's due to design rather than lack of a halo car.

maximum_sarge > Tim (Fractal Footwork) 05/06/2014 at 17:36

0

Kill the Wrangler and see what happens to Jeep... The answer is halo cars matter, but only to the extent that they represent and fulfill the brand ideals.

Humpty-D > Tim (Fractal Footwork) 05/06/2014 at 17:46

2

Interesting analysis. What cars exactly were looked at specifically for this article? I skimmed twice but couldn't find anything.

Not to sound too skeptical regarding this analysis you did, but how do you know that improvements in sales couldn't be attributed to other variables such as brand design, consumer ratings (reliability and quality), or even poor offerings from competing manufacturers? Is there a standard growth rate for sales that you compare these figures to that provide a basis?

Sebastien > Tim (Fractal Footwork) 05/06/2014 at 17:50

0

I think they pay off. For example, I thought McLaren was cool and all, but after the launch of the P1 I think they are as cool if not cooler than Ferrari (like the P13)

MooseKnuckles > Tim (Fractal Footwork) 05/06/2014 at 17:50

0

TL;DR

Based on not reading it, sorry, I'd say it worked for Audi. The CTS-V also likely helped Caddy a lot with it's image. And out of all Vettes, the C7 has turned a ton of attention to the brand that would have otherwise stuck up their nose at a Corvette, even though that nose-sticker-upper-person is only an enthusiast who drives a Tercel and not an owner of a Porsche or other C7 runner-up ;)

ABoostedRedFox > Tim (Fractal Footwork) 05/06/2014 at 17:52

0

As someone who had the option of writing anything for their final math paper in college...and who ultimately wrote about forced induction and calculating values of air both in vacuum, and in boost, i commend you sir. (and congrats on soon to be graduated as well)

My housemate who was also an business/econ member ended up writing his thesis (which i think ended up being somewhere around 100 pages cover to cover) wrote about Tesla and plans for future alternative energy in the automotive segment.

He wasnt a car guy by any stretch, but he asked to interview me and ultimately the thesis turned out damn good (because of my quotes and insights of course)

BonafideSupraman > Humpty-D 05/06/2014 at 18:00

0

I too am interested to know which brands were considered in this analysis, and which models were considered to be the eco halo cars and the performance halo cars.

HemiJKU > Desu-San-Desu 05/06/2014 at 18:02

0

I don't even play Halo but this made me laugh.

Bwahahahaha > Tim (Fractal Footwork) 05/06/2014 at 18:05

3

I would highly recommend investing the time to learn Stata or R if you plan to do more of this kind of research in the future. It may seem daunting at first, but it's worth it.

These programs will also make it much easier to identify the aforementioned "smelly smells."

Drakkon- Most Glorious and Upright Person of Genius > Tim (Fractal Footwork) 05/06/2014 at 18:08

0

Now! Racing programs as related to brand. I really don't think I want to see those results.

Clay...Bill Clay > Tim (Fractal Footwork) 05/06/2014 at 18:25

0

Yes, this is all well and good, but the people are demanding to know two things. One, did you induce stationarity into your time series? And two, which lag-length test did you use to select the optimal lag length? If you answer anything but Akaike, you're dead to me.

NJAnon2 > Tim (Fractal Footwork) 05/06/2014 at 18:30

0

Does it justify their development? Yes, when they are telling us NEW stuff is coming. It doesn't justify development when they are making more halo cars than actual cars. And they [car makers] do it all of the time. From the small market companies to the mass market.

CaptainStu > Tim (Fractal Footwork) 05/06/2014 at 19:02

0

TL;DR but to answer the headline when I see a Versa (or even Altima) I don't think GT-R. When I see a Dart I don't think Viper. It's not working but God bless em for trying. However when I see a WRX I do think that used to be an Impreza!

Time Traveler > Tim (Fractal Footwork) 05/06/2014 at 19:11

1

For those who didn't live at that time, the Plymouth Prowler was the halo car for the PT Cruiser. The Prowler was an instant classic and the PT Cruiser (I don't car what you think of it) was a big sales success. It ushered in the "Retro" craze that still lives on in many of todays most desired machines.

dariof01 > Tim (Fractal Footwork) 05/06/2014 at 19:20

0

nmalinoski > Tim (Fractal Footwork) 05/06/2014 at 19:28

0

tl;dr. The LFA does not make me want to buy a Toyota product. Subaru's halo car, on the other hand, does make me want to buy more Subarus (It helps that their halo car is <$40k USD).

Reckoning Day > Tim (Fractal Footwork) 05/06/2014 at 19:58

0

But of course...

Sullivanish > Tim (Fractal Footwork) 05/06/2014 at 20:06

0

Fantastic work.

Tim (Fractal Footwork) > PatBateman 05/06/2014 at 20:39

0

It's the standard linear regression used in Econometrics. I just added the variables I needed.

Tim (Fractal Footwork) > Humpty-D 05/06/2014 at 20:48

0

Oh but they can be attributed to other variables! You're absolutely right.

The data I used was very limited as there is almost no way in a 6 month course to gather every control variable out there. Not to mention it is not even available.

I definitely should have added in more controls, and that probably would have increased the confidence of the lagged dummy variables, but there is just no accessible data out there as manufactures want to keep their numbers relatively secret.

My conclusion doesn't make any outlandish assumptions off the regression that isn't supported by a level of confidence.

Blakkar > Tim (Fractal Footwork) 05/06/2014 at 20:58

0

This was probably not the best picture for the title. A "halo" car made by a company that makes nothing BUT halo cars. A halo car is car that more or less stand head a shoulders above the rest in a line up, to serve as the companies, tech demo, technological spear head and/or preview of what is heading into usually much less expensive models.

A $1.2M car the "trickles" down into $300K-$600K cars that are made within a year of the "halo" car (BTW all cars almost NO ONE outside of "Richistan" is ever going to see in person except at a car show), doesn't really engender caring about the subject, meaning, or intent of the article.

Better examples, to my mind, would be the Corvette or GT-R, as they do or at least seen a being influential on the rest of the companies line-up(s), which consists of far less expensive and usually much less or very differently capable vehicles. Other examples may include the Audi R8, SRT/DODGE Viper, AMG SLS, Jaguar F-type, among others.

A halo car from a company of halo cars, is not really a very meaningful halo car.

Axial > Tim (Fractal Footwork) 05/06/2014 at 21:10

3

My theory is that a halo car is most effective if you closely relate your bread and butter vehicles to it visually and technologically. Let's say you begin a new initiative. You simultaneously develop the halo car and the regular cars. The halo car uses the purest, most radical, and highest-performing versions of your aesthetics and engineering. You release the halo car first. It is well received, and even the less affluent want one. A short time later, you unveil the regular cars using watered down versions of what you find in the halo car. Now you've got people lining up to buy your cars because they want to be affiliated with your awesome halo car.

You've now created a brand identity. What's nice about this kind of development cycle is that you can use the halo car to herald each new generation of automobiles from your brand. You aren't compelled to follow traditional aesthetics for your halo car (a la Corvette, Porsche, etc.) because, by tradition, your halo car every time is supposed to be something new and fresh and your mainstream automobiles must follow its lead.

This kind of strategy is, however, heavily dependent on marketing. You'll have to really do your engineering homework to be able to leverage the press to your advantage. You might also have to spend a short while eating costs until you can grow your customer base.

But yeah. That's how I would use a halo car. Simply using it to show that your brand can create something amazing without also immediately relating it to your brand's bread-winners in more meaningful ways is an inefficient use of material; it's wasted potential.

GhostZ > Tim (Fractal Footwork) 05/06/2014 at 21:29

1

As an Econ Major, you make me cry tears of joy. This is wonderful. I have no other substantive comments I can make, you said it all.

Tim (Fractal Footwork) > Axial 05/06/2014 at 21:34

0

I would agree. Hence the whole Attribution Theory section.

Axial > Tim (Fractal Footwork) 05/06/2014 at 21:48

1

I confess I skimmed through. I'm current having my head jammed full of statistics and probability and regressions, etc. at school, so my tolerance for it in my down-time is severely diminished.

I just thought I'd extrapolate on that last paragraph under "Theory." Basically, car sales are Inception: you have to make the customer think that the idea of purchasing the car came from his own spontaneous thought and was not the result of you meddling around in his head with marketing speak. In practical terms, that usually means crafting a product good enough to sell itself.

avi24 > Tim (Fractal Footwork) 05/06/2014 at 22:20

0

Enjoyed the article and the findings. I remember studying this as an undergraduate (the economics regression - not the halo cars) and I find your work well written and well done.

Kudos.

laurence37 > Tim (Fractal Footwork) 05/06/2014 at 22:55

0

I think an alternate way to think about this is that profitable car companies that are gaining market share can afford to make performance halo cars. I think Nissan's financial success and growing market share has more to do with them making the GT-R than the GT-R contributes to Nissan's market share.

-Tom- > Tim (Fractal Footwork) 05/06/2014 at 23:38

0

A halo Viper didnt lure people into showrooms and entice them to buy Neons with the promise of the Neons being little Vipers....no, it was a sanity exercise for the engineers that gave everyone a little hope. Its also something that while it isnt a huge seller, gives people out in the world a smile as they see it going down the road. These cars arent designed to be sold by the millions or to trick consumers, they are made to make everyone smile just a little bit more...

NickD > Tim (Fractal Footwork) 05/06/2014 at 23:53

0

So essentially maybe,sometimes,not always

Stef Schrader > Tim (Fractal Footwork) 05/07/2014 at 00:16

0

Holy crap, nice piece.

keepingitreal44 > Tim (Fractal Footwork) 05/07/2014 at 00:28

0

offcourse they do, nobody buys a CLA250 because it's a bargain, they buy it because of the benz halo cars and want to be associated...kind of like joining a gang and want to feel like they belong to a particular group..The CLA is the cheapest price of admission to the popular group so to speak.....

The S-Class helps sell thousands of CLA's and that is something Lincoln/Cadillac/Acura to name a few don't understand, but Hyundai/Kia are following on the recipe.

leobrioschi > Tim (Fractal Footwork) 05/07/2014 at 00:52

0

I find it a bit odd that only "halos" were used as "betas". What about other important variables(missing those can imply a Type I error, like explained at: http://www.mathworks.com/help/econ/exam… )... (break cause Kinja don't want me to type what i want)
Have you tried simple ARIMA models? is there not a MA or AR relationship in this theme?
Have you tested for Autocorrelation on the residuals?

Can i read your paper?

sorry for being bitchy, my fellow Economist!
wish i had a capstone project that is so cool (not really, i got it easy 'cause i used my former research with my Professor, i didn't like the theme but i had most of the data crunching part done!)

Tim (Fractal Footwork) > leobrioschi 05/07/2014 at 00:59

0

I understand, I just didn't have access to data that I would have liked to use, nor the time to implement it.

LeTronique > Tim (Fractal Footwork) 05/07/2014 at 01:03

0

Damn! What did you use to run your analyses? This is a hell of a lot more complicated than Undergraduate SPSS. hahaha!

meatatarian > Tim (Fractal Footwork) 05/07/2014 at 01:28

0

SPSS was my sheeeit

xfgmn > justregisteralready 05/07/2014 at 01:31

0

Just did the same to your comment. Kudos to you for putting all that much effor into it. I'm sure it was a good read.

boxrocket > Axial 05/07/2014 at 01:32

1

Yes and no: Consider the Ford GT versus the Audi R8. Both good supercars in their own right, and easily considered halo cars for their brands, but didn't really do much "trickling-down" to the lower cars, if any at all. The Ford, other than sharing the basic engine with some Ford trucks and Mustang, was an exercise to bring nostalgia, pride, and futuristic hope to a brand that was trying to rebuild itself, which is now once again one of the largest automotive manufacturers by volume in the world. The Audi, on the other hand, was/is, essentially, a cheaper but more comfortable iteration of VAG's Lamborghini Gallardo, even sharing the powertrain (though the engine was also used in one of the S or RS models, I don't recall which), seemingly "just because" they needed another vehicle to show that performance and expensive vehicles are in their wheelhouse, and they're no longer just gussied-up VWs, so while it does somewhat accomplish its mission, it does so at the expense of another in-house brand - granted that few would be cross-shopping - just to seemingly prove a point. Audi has also grown in volume and apparent quality, and make quite remarkable vehicles (I drove two S4s today on the highway and was impressed with their power and acceleration at highway speeds, for instance), but few Audi-philes seem to point to the R8 as an inspirational vehicle, rather they seem to favor the bread-and-butter Quattro models and the performance S and RS models. In contrast, folks still fondly remember the Ford GT, and when new Shelby Mustangs come out, the two are compared (and that an almost ten-year-old car with "only" 550hp can do so well against a 2014 GT500 with 662hp is most impressive). The R8, OTOH, seems almost forgotten.

AngryLibrarian > Tim (Fractal Footwork) 05/07/2014 at 01:36

0

This is a great read, but I think it actually confirms what we already know. "We, the people who love cars, do not design them, buy them, or produce top performing models because we need to impress other people. We do it because we love cars." -AngryLibrarian, 2014, Jalopnik forums.

Axial > boxrocket 05/07/2014 at 02:24

0

While I do agree that pride in a brand is one of the benefits of a halo car, I am not sure I agree that the Ford GT had much of a halo effect. It's fun to talk about but most non-car people don't even know what it is. Even for those who do, it exists on a plane too far apart from the rest of Ford for the brand to benefit from its effects (there's that wasted potential). The Mustang has a far greater impact on the notion of pride in Ford's brand and actually is their halo car. When new Shelby Mustangs come out, I see it compared to the Corvette more often than any other car (which is an asinine comparison, but I digress).

As for the R8, it features little halo effect because it is a culmination of everything going on in the lesser cars. You can't really make a halo car via trickle-up effect. Trickle up is what you do when you've got money because business is good and you want to go out on a limb and do something neat without breaking the bank. You'll be hard pressed to make such a vehicle out as a halo car.

Charles Moseley > Tim (Fractal Footwork) 05/07/2014 at 06:27

0

In order to get a passing grade in my Statistics course at grad school I promised my professor to never claim to understand statistics. Therefor all I can say is 'this seems cool'.

You probably won't like me > Tim (Fractal Footwork) 05/07/2014 at 07:11

0

Two words about the tables. Significant figures. Yikes.

Chachmin > Tim (Fractal Footwork) 05/07/2014 at 07:42

0

Well, I have never regarded econometric methods as a magic wand that can yield some valuable results on its own, as long as in every regression there might be important factors that you leave outside the model, because: a) cannot estimate, b) cannot estimate properly/precisely enough, c) don't know about or deliberately ignore just because they seem insignificant, d) they are so obscure that nobody knows about them (like people with blonde hair may be significantly more likely to purchase blue crayons when it is raining (compared to other customers purchasing blue crayons when it's raining) for NO obvious reason). But wait, I digress.

In fact, I really enjoyed the author's insightful approach and analysis of previous research - which is very informative, too!

The only comment I'd like to make is that the author analyzed data on cars selling in the U.S. from 1961 to 2013. Unlike fast cars, the popularity of green cars in the U.S. must have remained somewhat moderate until recent decades (late nineties or even noughties (though it was quite a competitive advantage due to the oil crises in 1973 and 1979)). Anyway, I guess it has to be taken into account to make results for green halo cars more representative.

tekkblade128 > Tim (Fractal Footwork) 05/07/2014 at 08:04

0

This car does

PardonMyFlemish16 > Tim (Fractal Footwork) 05/07/2014 at 08:41

0

Halo cars don't work in any realm.... performance, luxury, green tech. Mainstream and low end buyers shop with blinders... some college grad chick doing a $299 lease on an IS250 doesn't know the IS-F or LFA ever existed. Nobody is buying Versa Notes because they share a brand name with a GT-R. This is one of many myths here that get propogated as truths.

iceman967 > Tim (Fractal Footwork) 05/07/2014 at 09:45

0

I have a stats final on monday, it is nice to fell like reading a Jalopnik article is actually me studying.

jimmy-buffett > Tim (Fractal Footwork) 05/07/2014 at 10:31

0

It would be nice if the company's halo car wasn't astronomically priced with a low fixed production count, case in point: the Nissan GT-R.

I'm a big Honda/Toyota guy and the Lexus LFA and proposed Acura NSX are both looking to be sales failures. Even today you can find one of the 500 LFA's ever built at dealerships in AutoTrader two years after they came out, there's a full page of them with under 1000 miles (and several with under 100 miles). The NSX seems to want to be a baby McLaren P1/Porsche 918 but with a power rating (equal to) and price point (roughly double) that is going to make it a loser compared to the GT-R.

Honestly, the GT-R has ruined the sub-$250K market. If you're buying a performance car in this price range and you're not getting the GT-R, it's because you care more about looks or badge than performance. Which brings me back to the NSX: at $200K with a V6 it will be a good performance competitor but overpriced compared to the Nissan. Why not bump the engine displacement down to 2-2.4L and make it the P1/918 version of the $30K Evo/STi crowd, at $70K? I'd actually buy that car, and it would be another impressive car for the price point (like the GT-R) rather than just another overpriced undersold victim of Godzilla.

milkandhoney > Tim (Fractal Footwork) 05/07/2014 at 10:52

0

You mentioned this was a capstone... at what university?

Humpty-D > Tim (Fractal Footwork) 05/07/2014 at 10:53

0

I completely understand the lack of data, excellent analysis either way!

Stephen Karlson > milkandhoney 05/08/2014 at 17:40

0

ARIMA models and serial correlation? Well-played, commenters. I'm not sure the undergraduate econometrics sequence at Northern Illinois covers that.