The Car Bundle
Two Truths and a Take, Season 1 Episode 32
There’s a lot you can criticize about cars and individual car ownership. But one criticism you shouldn’t make blindly is “Think about how much of the car you don’t use.”
It’s true that not every driver or every trip needs the horsepower of an ICE or a top speed of 65+ MPH. Not every trip needs four seats, or even two. Not every trip needs climate control or cargo space. But it doesn’t matter, because cars are a bundle.
A car is like cable TV. There are some features of my car that I don’t really need. And there are some that you don’t really need, either. But we both buy them anyway because both the logistics and the economics of buying all of them together are so much more compelling than if we each had to individually shop for only what we needed in a vehicle. Today, we’re going to talk about why.
Why Bundles Work
Bundle economics can be tricky to understand - how is paying for stuff you don’t want a good deal? Fortunately, Chris Dixon already did, really well:
How bundling benefits sellers and buyers | Chris Dixon (Word for word, one of the most useful things I’ve ever found on the internet, 7 years later.)
Bundles naturally occur whenever you have a bunch of related product offerings or capabilities, and a wide range of willingness-to-pay among potential customers. If you offer each product individually, and price that product in order to maximize revenue, the revenue-maximizing strategy can often be to charge a high price, and monetize a relatively small number of customers. It’s revenue-maximizing, but it’s not ideal: a lot of people miss out on the product entirely (they’re "deadweight loss” to the seller, as they generate zero revenue) and the people who do buy the product are paying close to the limit of what they’d be willing to pay (so there’s not much consumer surplus).
If you bundle the products together, on the other hand, a funny kind of economic magic happens. Everyone ends up paying an approximately fair price for the one product that they were going to buy no matter what, and then gets a very good deal on everything else. The consumer surplus from all of the different products you bundle together adds up faster than the cost of the bundle - the “free lunch” comes at the expense of deadweight loss. The net result is that sellers make more money, and buyers get a better deal: you pay somewhat more, but get way more, than if you were shopping individually and only bought your priorities.
Cars bundle several jobs together
Cars are also a place where a lot of different features people want are bundled together into a single offering. If you’ve ever bought a new car, you’ve probably faced down the endless lists of extra add-on features in cars (bigger rims, heated seats, LED lights, whatever) are always, always sold in these “packs” where 5 or 6 totally unrelated things are offered together in a bundle. Now you know why.
But at a broader level, it’s useful to think of the car itself as a bundle. The value proposition of a car is made out of a few different core features, like “it can get you to work”, or “It can take you shopping.” Buyers have a wide range of willingness-to-pay for those functions: not everybody wants everything. Within those core functions, there are a lot of possible sub-features you might care about. Some people care about horsepower, some people care about trunk space, some people care about style, some people care about safety.
There’s a good reason why SUVs have become so popular: they’re the everything bundle. No one needs every feature of an SUV. But SUV buyers are quite satisfied with their cars, because they’re getting so much car. The bundle math works out in favour of everybody buying a vehicle that’s capable of every different need, even ones you don’t really have. (Again, read Chris’s post for why.)
Once you have a car, it makes sense for you to use it (in all of its featured glory) all the time, since you’ve already paid for it. Economically speaking, dragging 2 tons of car along with you just to go get groceries is a little bit like paying $5 for single day of cable TV even though all you watched that day was two minutes of the weather channel. From the point of view of that specific trip, it’s a giant waste. But from a bundle economics point of view, it’s still one of the greatest free lunches you’re ever gonna get.
Even if you put aside the some of the more emotional or expressive aspects of car ownership (which are powerful in their own right), car ownership is a compelling deal because it bundles together four core jobs, and practically all car owners use them for at least two of these things: commuting, shopping, kids, and recreation. If one or two of those tasks are essential for you, the car will hold pretty strong as an indispensable possession for most people, at least in North America.
That’s why ditching car ownership is going to be really unattractive for a lot of people - no matter how attractive you make the alternatives. Unless you can replace all of the important jobs that a car does for you, all at once, then competing against the car means competing against free. Actually, it’s worse than that - it means competing against free and nice. Bundle economics (and also ego issues) are powerful enough that it’s pretty rare to see people downgrade their cars, even if their car requirements have gone way down (like they had kids go off to college). Once you go SUV, you don’t go back.
Threats to the car bundle
What would it take for this bundle to come apart? In order to “steal” trips away from the car (whose marginal cost of use for most trips is basically free), you need an alternative that is both more compelling than driving, and that is cost-competitive with free. For the first two, real things are happening:
Commuting: The first major thing we use our car for is getting to work. And there are two big trends here that matter.
The first one is the recent trend of a lot of office jobs heading back downtown, and back into major cities, as traffic in those cities gets progressively worse. If your job is actually reachable by mass transit, walking or biking, then you can potentially unbundle that task from the car. That’s a big one.
The second big trend that matters here is remote work. There’s a solid chance that remote work will become a big thing over the next twenty years - not “big" like it is today (growing and relevant in tech but that’s kinda it) but actually big, at the workforce level, for all kinds of knowledge workers in every different field. In this case, you’ve replaced commuting with virtual presence, which has its own set of costs, but they’re coming down rapidly.
Shopping: The second big thing we use our cars for is getting stuff. We shop for household stuff, we shop for food, we shop for all kinds of things, and we use our cars to get there and back. Here there’s been a major change recently, which is that the rise of e-commerce has turned the last mile into someone else’s job. The stuff is still being driven, just not by you.
There’s a cost to all of this, of course: shipping and delivery costs. Or, at least, there used to be. But all that driving is getting re-assembled into new bundles, like Amazon Prime, where “free” next day shipping is now simply something that we expect is incorporated into the purchase price, no different from the costs of fulfillment or warehouse storage. Amazon is still only a single digit percentage of American retail, but overall trend is undeniable: the internet means “Get the stuff to me” is increasingly done as a service for you.
Four corners, five delivery vans: just a regular 2 pm on a Tuesday.
Imminent, cheaper form factors like drones and cargo AVs will accelerate this trend, but they aren’t necessary. Software and the internet made delivery drivers so productive and omnipresent that the marginal cost of giving them one more package to deliver became, for the most part, “low enough.” (Admittedly, a lot of this is possible because the true cost of delivery is actually offloaded as a negative externality on the surroundings: delivery trucks blocking lanes and clogging streets is a normal frustration now.)
1.5 million packages a day: the internet brings chaos to NY streets | Matthew Haag & Winnie Hu, NYT
If you drag this trend forward a decade or two, it’s reasonable to think that a meaningful percentage of people are going to have the majority of their retail purchases delivered to them. Most American households have an Amazon Prime account. They’re already paying for the cost of delivery (of everything) in a new bundle. I think people are going to be surprised how much of commerce is going to shift in this direction, especially as a new cohort of eCommerce-first kids grows up and starts their own households.
Expanding the car bundle
So let’s be really futuristic for a second, and fully subtract those two big use cases from the car bundle value proposition. What’s left? Two big ones: kids and recreation.
Kids and recreation will be to the car bundle as live sports is to the TV bundle: the enduring, critical function that keeps the bundle together for a long time. It’s not surprising to me that "kids and recreation", as a category, probably comprise 70% of the content of car ads I see on TV. (The other 30% are mostly pickup trucks hauling fake dirt around national parks.)
Kids and recreation are also going to be the two hardest categories of driving to displace, seeing as they’re the most emotional and non-utilitarian aspects of car ownership. For most North Americans, anyway, you’ll probably keep driving your kids around in your own car long into a future even if it’s filled with driverless cars and on-demand services and who knows what. That use case is strong enough, in fact, that it could serve as a really good anchor for a new car bundle that’s just as strong as the first one, but filled with a totally different set of utilities and use cases - but organized more around entertainment and consumption than around utility. What could this bundle look like?
I think that smart car companies, if they see this the same way I do, will keep adding high-quality media people to their boards and management teams. These people understand bundling better than anyone. And I think they understand American car buyers, who through their revealed preferences are increasingly selecting expensive, recreationally-oriented SUVs as their vehicles of choice. Is there a bundle, like a cable bundle, where including an SUV actually makes sense?
I was initially thinking subscriptions like Disney+ would be no-brainers, but on further thought I don’t think that’s actually the right approach. More or less everyone is going to get Disney+ regardless of whether or not it comes with their car, so there’s not much deadweight loss ought of which you can go look for economic free lunch. Plus it’s too small an expense to matter.
Good bundles are made out of a collection of product offerings, each of which only a small percentage of people would actually buy today, but which do genuinely have mass appeal to a lot of people if the price got low enough. They also have be expensive enough to be a real factor in purchasing a car. $10/month media subscriptions aren’t going to cut it.
I have an idea here that’s radical enough to qualify as a hot take but I also think makes total sense. To me, there’s one massive but surprising candidate that fulfils all of these criteria: 1. kids / recreation oriented; 2. fragmented enough for bundles to work, but which appeals to everybody; and 3. A large enough market and expensive enough purchase for the combined offering to be really compelling. Vacations.
If you’re a car OEM, I think there’s an incredible opportunity over the next decade to build a new bundle around recreational travel and leisure, where you can put together a really compelling bundle offering full of products from other companies that are complementary but not competitive: hotel and resort chains, cruise lines, kids programs, maybe even airlines. The amount of deadweight loss you can draw from in order to make the bundle magic work is almost infinite - travel is super fragmented, and everyone would love to go on more (or better) vacations if they were cheaper. Plus, people spend a lot of money on family vacations. Probably as much or more as on their cars, when all’s said and done. Unlike Disney+ or its kin, the financial toll of family vacations is large enough to really matter in a bundle like this, and not simply get dominated by the cost of the car.
Bundling a large, material purchase like a car together with occasional experiences like travel might seem like a strange combo. But I think it could make a lot of sense. For the right buyer (and it’s a large percentage of them!) it’s really on brand, especially as far as kids are concerned. It makes a ton of sense to link together the idea of exploring / adventuring / family time (which vacations ring up in spades) to the kinds of emotions you want people associating with your car, but before and after a purchase. For car companies, bundling in vacations hits the right notes.
But why would travel and vacation businesses go along with this? Because they are in huge trouble if they can’t work out a major new customer acquisition channel that isn’t Google or other forms of pay to play user acquisition. They need to get creative with finding a new distribution partner that are heavyweights and are everywhere. Cars, and car advertising, and car dealerships, might be the perfect place to go. Car companies know how to sell a lifestyle.
What would this car bundle compete with? I don’t think there will be too many people out there thinking, “Well, I’m not sure whether or not to buy a car. But I’ll buy it if it comes with all these sweet vacation products!” It’s not competing against non-car ownership. I think what it competes against is car downsizing. This bundle isn’t how you sell a car versus nothing, but it’s how you keep selling more and more expensive cars.
An SUV bundled with, say, Universal Studios, Norwegian Cruise Lines, and Southwest Airlines as “anchor tenants” and then a big list of ancillary travel services only a tiny number of people ever deliberately seek out, but a lot of people would use if it were offered to them for “free”, would be a pretty powerful product offering, compared to the more sensible option of buying, you know, a regular car.
Over ten years or so, you could build this into an interesting reinvention of how cars are marketed and sold that plays to their enduring strengths, builds on their lifestyle branding, and creates a whole lot of value for a lot of different buyers and sellers. And the bundle magic will really work: customers who buy it will pay more than they need to, but they’ll get a lot more than they would have otherwise. Is it irresponsible? Yeah, probably, for most people. Will it work? You bet it’ll work. Especially if the car part of the bundle helps get the whole thing financed for you.
It’ll be largely, but not entirely, a biz dev exercise. There’s room in here for a startup to invent a really compelling form factor to get a wedge in the door, and I bet you it starts on the vacation side of things - specifically targeted at families with kids, and developed with one or two car OEMs in mind from the very beginning. I’m not sure what it looks like, but I bet you it could exist! If you have ideas for how this might look, please drop me a line, I’m all ears.
Speaking of bundling, this interview with John Malone is a must-read for everybody. It’s so good. (Here’s the video.)
A great tweetstorm from Mike (@nonGaap) on the always fascinating food delivery wars
Doordash illegally charges state tax on food orders, class says | Perry Cooper, Bloomberg Law
Food delivery services taxed by duty question | WSJ
And another fantastic history lesson from @mcclure111 on how Sun Microsystems fully invented - and implemented! - cloud computing 15 years ago.
Finally, just for fun: here’s an outlandish but I think actually kinda plausible prediction on the American primary elections coming up:
Pete Buttigieg wins Iowa. The media embraces him the clear national front runner. Shortly afterwards, a TikTok meme roasting him goes so viral and is so devastating that, when Bernie / Warren win the next couple states, the media seizes on this meme as a story, and finally “gets” what TikTok is about. It’s a meme they actually understand, and it has real consequences that serve them well (by creating more primary drama.) So it becomes its own news cycle.
The TikTok meme itself has no impact on anybody’s vote, but the resulting reflexive media cycle does: it becomes Dean Scream 2.0, ultimately sending Buttigieg into a media death spiral that they can’t do much about. Meanwhile, all of this attention makes TikTok start to jump the shark: with everyone talking about it on TV all day, parents start asking their kids about it too much, and it starts to slide towards un-coolness. So the Mayor Pete meme, in becoming an enormous triumph, seeds the undoing both of Mayor Pete’s campaign and of TikTok itself.
That’s my theory! Timestamp it November 19.
Have a great week,
Alex