In a previous post I looked at which features are important for a laptop’s price. This post uses the same data (a sample of the 101 most popular laptops from Amazon) and with it I try to find out which companies can sell their laptops at a premium compared to similar laptops of different brands.
I identify three brands, which stand out of the crowd. So which brands are so strong that they can afford to charge a significantly higher price than their competitors for similar set-ups? Read on to find out.
Let me start out by summarizing my research: I used the same dataset and the same basic model as in my first post on laptop pricing. The only difference is that I added dummies so that each brand’s seperate effect can be identified (11 dummies for 12 brands). Some brands didn’t have many laptops in the sample (like Gateway or Samsung), so it may be worth reestimating this model with a different sample. I may do this at a later stage, but for now I wanted to get something else out of this set of data.
In the regression I took Acer to be the default brand (for alphabetical reasons). So the coefficients show by what percentage each brand is more/less expensive than Acer, while controlling for screen size, CPU, memory, GPU, hard drive and the presence of a touchscreen. The results say that 8 out of the 11 remaining brands do not charge a significantly different price than Acer does. From here on out, I will refer to this group of brands as “average brands”.
It appears thus that there are only three brands that can charge a price premium based solely on their names. One of them is of course Apple, but it may come as somewhat of a surprise that Apple is not the one with the largest price premium.
Alienware, a high-end brand (and apparently a subsidiary of Dell) with only 490 employees can charge even higher prices than Apple. Remember that we’re controlling for hardware characteristics, so this is pure brand name premium. An Alienware laptop is as much as 94% more expensive in our sample than a similar laptop of an average brand. The confidence interval says that in any case, Alienware will probably be able to charge a premium of 58% to 130%.
Apple, coming in second, can price their laptops 66% above an average brand. In the regression model presented in the previous post on laptop pricing, this value was only ~50% because then Apple was considered against all other brands (including for instance Alienware), now it is considered only against average brands (see definition above). Apple’s confidence interval starts at 27% and goes all the way up to 104%.
The third brand that can charge a price premium is a surprise to me, it is ASUS. ASUS to me personally was always the brand with the best price-quality ratio, so it’s interesting to see that they’re actually more expensive than the rest. Maybe my experience was just an illusion (I mostly shopped in the low-end market, where most of ASUS’s competitors like Sony or Toshiba weren’t even present), or maybe ASUS has started becoming more upmarket recently given its success (my experience is several years old).
In any case, in the sample ASUS could command a price premium of 29% over average brands. The confidence interval says, however, that the true value can lie anywhere between 2% and 56%.
The rest of the brands cannot / do not charge significantly different prices based purely on brand name. If I were to ignore the p-values and set up a ranking of brands based on price premia, then this would be the following:
1. Alienware ***
2. Apple **
3. ASUS *
***, **, * = significantly different from Acer at the 0.1%, 1% and 5% levels.
Please note again, that only the first three brands are statistically significantly different from Acer. Therefore, I would treat the rest of the ranking with caution. For all we know, the remaining nine brands can fill up places 4-12 in any order, or more likely they could all tie for 4th place.
One additional thing to consider is that Alienware’s and possibly Apple’s price premia can be overestimated by the regression. Both of these brands (especially Alienware) only enter the market at its high-end. It seems, however, plausible that an additional gigabyte of memory will not cost the same when you’re upgrading from 4GB to 5GB as when you’re upgrading from 16GB to 17GB. This could be because of technological constraints and because of economic constraints.
I’ll leave the technological constraints to the experts, but let’s consider economic constraints for a minute. In the high-end market demand is probably more inelastic, therefore suppliers can charge higher prices. This inelasticity may come from the fact that someone who needs a high-end laptop probably needs it for important reasons (work) or is quite rich. Both of these would indicate that the consumer’s demand is inelastic, i.e. they will buy the laptop no matter how much it costs. There is also fewer suppliers at the high-end of the market. Even though they’re competing for fewer consumers, fewer suppliers could mean less competition and thus higher prices.
Consider the following graph:
The black line shows what the regression would predict: better features lead to higher prices but in a linear fashion (i.e. adding that additional gigabyte of memory costs the same at all levels). The red curve shows the situation described above: that costs/prices can skyrocket after a certain limit (when demand is inelastic enough). If a brand sells most of its products where the red curve is above the black line, then its brand name premium is likely to be overestimated by the regression provided that the hypothesis of economic and technological constraints is true.
For this reason, I would say that Alienware’s (whose cheapest laptop in the sample is a 17.3-incher with 8 GB memory, a 2.4 Ghz Core i7 CPU and a mid-range GPU) brand name premium is probably overestimated by the model. The same may not be true of Apple, at least not to the same extent. Therefore, it is possible after all that Apple has the highest price premium or at least that it sits in the #2 position closer to Alienware than indicated above.