I guess it’s serendipity that this post showed up over at Yglesias’s blog this morning. It discusses a paper that argues that deficits actually lead to more demand for government, as shifting costs to the future lowers the public’s sense of the cost of government services, prompting them to consume more. This doesn’t seem very controversial to me (or at least I’m not qualified to say anything interesting about it).
But just yesterday I was thinking about something very similar in the context of the current storm over Twitter. Not so much because of the specific question of how much Twitter client variety we’ll see, but the larger question of the firm beginning to make decisions that will help it generate revenue rather than decisions that will help it generate user adoption.
And it occurred to me that we basically debt finance our social networks. There are a few exceptions, but in general nobody’s willing to pay for social experiences on the web. So instead we’ve arrived at a model where open registration, open APIs and open bars at SXSW are used to attract users. Then, once network effects have given the experience value, and so long as transition costs remain inconveniently high, value can be extracted from those users. Lots of people tell lots of stories about how this can be a positive-sum interaction (usually fairytales about how valuable anonymous usage data is, or how people will suddenly start enjoying the experience of looking at ads if the sales pitch is sufficiently personalized). But in practice it seems to always be the case that user experience is degraded in some way.
This isn’t a particularly great insight, I suppose: that you can goose demand if you shift costs into the future. Discount rates exist; it’s why economists fret about declines in consumer borrowing. It’s why people still smoke cigarettes.
I have an aesthetic aversion to it — a general sense that things would be better if costs were presented up front. That it would be preferable for Twitter to be a protocol like email rather than a service like Facebook. Maybe even that it would be more moral for them to be. That the costs associated with transitioning between social networks every, what, five years are high enough to be worth complaining about.*
But this is probably is just a meaningless prejudice on part. Twitter wouldn’t have become Twitter if those costs had been apparent from the start. To wish it could be otherwise is just to say that you want foolish angel investors to buy you toys for free**. The service will probably remain fine for a long time, and then it’ll get worse, and we’ll all bitch about it and eventually something else will come along.
I just hope that when Google finally decides it needs to make some money off Gmail that they have the class to ask me for it to my face instead of creating some privacy-violating Rube Goldberg device by which other corporations agree to exchange money for making my life very slightly worse, as if I was plugged into the Matrix to power cybernetic tyrants with my mild irritation.
* Whatever this inclination is, it’s closely related to the sense I have that mobile vendors shouldn’t use contracts to subsidize handsets, and that cable TV should be a la carte, and that bandwidth should be billed on a metered basis. In all cases these “innovations” probably facilitate commercial activity that’s desirable, but I’d be happy enough to see it disappear simply to avoid the sense that people are being duped. I guess it’s just pique.
** To be clear, I’m in favor of this.
Good post, and I agree in general.
This is a nitpick, but I’m confused about why you chose metered bandwidth and a la carte cable as examples. Those aren’t trade-offs between paying now or paying later, they’re about economizing on transaction costs. Do you think McDonalds should charge for ketchup packets and bathroom visits?
Well, I chose them precisely because of their fuzziness: I agree that they aren’t examples of the same cost-shifting. But they *are* examples of the same aesthetic preference that I have for certain kinds of granular, contract-free transactions.
(As for whether those examples are (still) about avoiding transaction costs: we’ll have to agree to disagree — I think they’re at least to some degree about obfuscation with the intent of tricking consumers into buying more than they otherwise would).
Oh, one other thing: it seems likely that the costs of running these services are low enough that companies could choose to take only modest profits, (through vanilla non-personalized advertising, say) and not turn evil. Craig’s List seems like an existence proof. The problem is that the investors who put up the money in the first place are never satisfied with only modest profits, so they push for more ambitious monetization strategies that ruin the user experience in the process.
So maybe what’s needed is a community of Mark Shuttleworth-style philanthropic angels that are willing to finance non-profit Facebook/Twitter/GMail/etc competitors in which they hold debt rather than equity. If the sites succeed they’d get their money back, but any subsequent upside would flow to users.
I also liked the post, but was jarred by the “metered bandwidth” analogy. You really think that’s a good idea? There’s no contract involved with my unmetered cable subscription, and, though some quick googling isn’t turning me up any studies, I feel pretty confident that any performed would show that people use a lot less bandwidth — and generate a lot less economic activity — when they feel like they have to watch what they do all the time. Similar to how people in DC are using crazy fewer bags to avoid a trivial 5 cent tax, cause now they pay attention (though in case such reduction is a good thing).
Yeah, I do think it’s a good idea. ISPs need a lever to control network use, right? Given that reality, your choices are between price mechanisms or regulation — the latter in the form of bandwidth caps, letters about terms of service violations, and other unappealing fiats from Comcast et al.
But it seems pretty clear that Americans prefer the all-you-can-eat-buffet (but the food sucks and we might decide to throw you out whenever) model for these kinds of purchases. And even if I got my way, there are of course fixed costs associated with Comcast maintaining a line to your house and billing you for it, even if you don’t use any bits in a particular month. So realistically we’re talking about some usage cap, after which overage fees kick in (hopefully with a friendly email letting you know you’ll be paying a few dollars more if you continue to seed that Ubuntu distro this month). And in fact I think that’s the model that’s emerging.
(I also think that metered pricing tends to make the service provider’s marginal expenses more transparent to consumers, which ideally drives competition and lower prices).
As an example of the kind of inefficiency I dislike, consider how your bill for mobile voice service has never gotten cheaper. This is true for most segments* of the market despite the fact that those vendors have been enjoying benefits from better technology, productivity, and paid-down infrastructure investments. Instead they’ve “given” the surplus to consumers in the form of minutes they don’t actually want or use and kept the money for themselves. This, incidentally, is Dirty Trick Number 1 when corporate shills write studies about how great for consumers the US telecom landscape is.
* My impression is that the prepaid mobile resellers really have brought their prices down, both because their model necessitates it and because the people they’re selling to are probably extra price sensitive.
I agree with you and I wish that there was a choice. The conventional wisdom is that no one will use a service if the costs are clear, so we aren’t even given the option.
Unfortunately, I think so few people agree with us that there’s not much hope for change. For example, the all-you-can-eat buffets are always packed when my mom drags me there despite the crappy food.
I’m impressed with your consistency. Quite a lot of people hate separate pricing in some contexts (airline fees, metered Internet access), but love the concept in other cases (a la carte cable television) and are neutral or mixed about others (subsidized handsets on contract).
Bundling creates winners and losers within different groups of consumers, and that partly (but not fully, I think) explains different reactions. Many people favor bundling when they’re a relative winner.
As regards mobile voice service, here’s the latest FCC report. Actually, the bill for mobile voice service for the overall market has gone down, ARPU from $47.23/month in 2004 to $36.98/month in 2008. The total ARPU hasn’t, because text and data has replaced that. I think it’s a dirty trick to exclude the part of the market with the most declining prices in order to argue that the non-price sensitive part of the market hasn’t decreased. I’d also argue that the increased voice minutes make it easier for more people to go without a landline, which then results in household savings on voice in one sense. (And caller pays makes it easier to get rid of a landline as well, as anyone who has called an international cellular phone and seen the different rates knows.)
That report also disputes your Dirty Trick Number 1 allegation by noting that Americans do end up using a lot more voice minutes than the rest of the world, so it’s not quite that people don’t want or use those added minutes. Perhaps you don’t, but perhaps it’s just like the behavior of people in other all-you-can-eat situations, and the majority of people are different than you.