Last week, I figured out that I am a part-time locust. Here’s how it happened.
I was picking the brain of a restauranteur for insight into things like Groupon. He confirmed what we all understand in the abstract: that these deals are terrible for the businesses that offer them; that they draw in nomadic deal hunters from a vast surrounding region who are unlikely to ever return; that most deal-hunters carefully ensure that they spend just the deal amount or slightly more; that a badly designed offer can bankrupt a small business.
He added one little factoid I did not know: offering a Groupon deal is by now so strongly associated with a desperate, dying restaurant that professional food critics tend to write off any restaurant that offers one without even trying it.
Yet, I’ve used (and continue to use) these services and don’t feel entirely terrible about doing so, or truly complicit in the depredations of Groupon. Why? It’s because, like most of the working class, I’ve developed a locust morality.
Thinking about locusts and the behavior of customers around services like Groupon, I’ve become convinced that the phrase “sharing economy” is mostly a case of putting lipstick on a pig. What we have here is a locust economy. Let me explain what that means.
I, Locust
Why locusts? Because I just learned a fascinating fact about them: they are not a separate species.
Locusts are the swarming phase of certain polymorphic grasshoppers that are normally solitary creatures, but under certain conditions (such as overcrowding, certain chemical trigger conditions and other subtleties I don’t get) they change their physical appearance, get a lot stronger, a lot more gregarious and aggressive, and form huge nomadic swarms. The picture above shows the two phases of the desert locust. At one time, they were thought to be distinct species.
When I first learned about this bit of science trivia last week, my immediate reaction was “wow, locusts are basically zombie grasshoppers!” But the more I thought about it, the more it struck me that this is a case of fact being stranger than fiction. Locusts are stranger and scarier than the zombies of our imagination.
And the more I thought about it, the more I also realized that I was a locust. Maybe 1/8 locust as a consumer, and probably 1/2 locust as an economic producer.
If you’ve ever participated in the “sharing economy” in any way, not just the worst-case act of offering or using a Groupon deal, you’re a locust too. Participation in anything from Airbnb to Zipcar (convenient A-Z phrase there) makes you at least partly a locust. It also does not matter whether you use a p2p resource base, one supplied by a small business sector, or one supplied by a big company. You’re a locust in all those cases.
I expect there are people out there who avoid operations of known dubious repute like Groupon, but the idea that this alone frees you of complicity in the locust economy is laughable. It is far bigger, and far harder to escape from than you think.
Zombies versus Locusts
For some time now, I’ve been fond of the idea that consumers are like zombies, and that during recessions, they are scarier than big capitalists (vampires). A few years ago, there was a vigorous online debate about the relationship between the state of the economy and the relative popularity of vampire and zombie movies.
Apparently, zombie movies are more popular during recessions.
It’s not hard to see why the analogy is so attractive. The identification of mindless consumerism with zombie behavior is very tempting. I offered my own related spin a few years back (The Gollum Effect).
But locusts bring a whole new level of detail, real-world precedent and seriousness to this class of metaphors. Because unlike fictional zombies, locusts aren’t the dumb instruments of a witch doctor or a mutant virus. They are actually smart and self-interested automatons, which makes them more scary.
In 19th century America, huge Rocky Mountain locust plagues were a feature of agricultural life. One swarm was recorded at 198,000 square miles (larger than California) and over 12 trillion insects, the “greatest concentration of animals ever speculatively guessed” according to the Guinness Book of World Records (via Wikipedia).
The economic devastation was big enough ($200 million between 1873 to 1877) to measurably impact the national economy during the plague years.
Then they suddenly and rather mysteriously went extinct. More on that later, let’s cover the basics first.
Locust Dynamics
I didn’t quite understand the entomology I read up on, during my few hours of browsing locust facts, but I came away with the amateur understanding that the reason for locust swarms is probably some mix of
- Opportunistic exploitation of a pattern of scarcity epochs punctuated by abundance spikes
- An element of predator satiation: a survival strategy based on overwhelming predators with numbers so that more of the species survives.
Locust swarms are also among the rare nomadic biological entities besides humans (nomadism is not a predictable pattern of movement, unlike migration,which is typically a sustainable pattern of movement through a resource landscape that’s not being devastated by the movement). They are nature’s rioting mobs, moving opportunistically from one store of food to another, without much concern for sustainability.
Whatever the biological details, the key point is that locusts devastate their foraging base.
Locust swarms don’t create new value. At a systemic level, the most charitable thing that can be said about them is that they efficiently strip mine value in a tyranny-of-the-biomass-majority way.
They out-compete other species through sheer numbers, and leave others to pick up the pieces as they return to their solitary, non-swarming grasshopper phase. In this case, human farmers. The collapse of locust swarms completes the cycle in a way we’ll get to.
Locust economies are built around 3-way markets: a swarming platform “organizer” player who efficiently disseminates information about transient, local resource surpluses, a locust species in dormant grasshopper mode, and a base for predation that exhibits a scarcity-abundance cycle.
So long as different locations are not synchronized, a locust market will usually have a surplus somewhere, even if it is a zero-sum or negative-sum market overall. Where that surplus comes from varies. In human farming, it is a natural consequence of the plant-harvest model.
Unfortunately, within the human world, it seems that the prey base is usually some sort of small business sector (either independent or franchisee chains). I will call this the Jeffersonian middle class, as in economic actors driven by the producerist values espoused by Thomas Jefferson (basically “small, local and independent”).
The war between the 1% and 99% seems to play out with the 1% and the 90% collaborating to prey on the 9% in the middle — the Jeffersonian middle class.
This is a very different class than the paycheck middle class, which has a superficially similar financial life. But during hard times, the paycheck middle class turns into the locust class in both production and consumption behaviors.
In a locust economy, the Jeffersonian middle class is a terrible place to be. It is no accident that the worst-hit victims of the locust plagues of the 19th century were small farmers living the Jeffersonian dream handed to them by the Homestead Act of 1862.
Abundance and Scarcity
Locust swarms are aggressively and energetically social. Remind you of any contemporary pattern of human behavior?
There seems to be an implicit holier-than-thou assumption among sharing economy evangelists that social sharing is primarily about virtuous behaviors like generosity, empathy, minding the planet, conserving resources, avoiding waste and so forth. Only secondarily do they see it as a zero-sum/negative-sum adaptation to recessionary conditions — Bruce Sterling’s favela chic.
They rarely think of it as a predatory behavior at all.
I haven’t yet worked out the details, but it seems to me that there is a curious interplay of scarcity and abundance in locust economics. On the one hand, there is local scarcity. Only so much wheat to eat locally, only so many local coupons offered by restaurants desperate for marketing leverage.
Once you’ve used up your coupon from the local paper, you have to pay full price. But if you can obtain news about another coupon in the neighboring town, the game changes.
Once locusts acquire an informed kind of market mobility through better discovery mechanisms, they can range over a much larger area of wheat fields or restaurants. You can continuously derive savings at the expense of other economic actors (wheat farmers or restaurant owners).
Once you increase your foraging range sufficiently, “local” means the smallest area within which you never have to pay full price.
So the abundance here is an illusion created on top of local scarcity via cheap discovery of transient local surpluses, artificially created by small Jeffersonian middle class actors hopelessly looking for leverage, and co-opted by swarming-platform owners.
Whether it is underutilized inventory of living space, coupons, parked cars or anything else, for most of the people, most of the time, if you have enough of the market navigation information, and are willing to travel a little further than you normally do, you can always find a deal.
The catch is of course, that for platform organizers to be profitable, they have to aggregate such slightly evil locust instincts and create locust swarms. No individual restaurant patron sets out on a 20-mile drive to a neighboring suburb with the malicious intent of helping bankrupt a restaurant there. Most sincerely believe they are just enjoying an exploratory adventure, and most would probably go back if they discovered something sufficiently unique and amazing via a Groupon deal.
But the way the aggregation model for the swarming platform works, the only incentives that can be offered are the ones that work on those least capable of hooking a returning customer through unique value. A truly unique restaurant will likely be well-known and constantly overbooked anyway, and have no need to offer a financially crippling deal. Even if a particular individual charitably decides to return to “give back” to a particular restaurant, there is a good chance the locust attack has put it out of business, making individual good intentions moot.
So the system can only create transient locust swarms out of individual slightly evil intentions. A locust market is one which looks more information-efficient than grasshopper markets, but really just has a pattern of information flow that favors different actors.
I suspect it is possible to design an incentive scheme that would actually be sustainable, but still offer customers some opportunity for consistent savings through nomadic swarming. Some sort of hybrid locust-grasshopper economy. But I’ll leave that question aside for now.
Software and Locusts
I’ve been thinking a lot lately about Marc Andreessen’s famous line, “software is eating everything.” The specific choice of the word “eating” is revealing.
Locusts also eat everything.
Well, almost everything.
What’s really going on is that software-enabled human locust swarms are eating everything they can access. Which generally means small business front-end layers wrapped around larger platforms. The locust swarms cannot actually take on true Big Industry unaided, for the most part. When Big Industry owns its own last mile (think McDonald’s) it is rarely stupid enough to offer up lunch for locusts.
Those who believe that the Internet-enabled economy is going to replace “hierarchies with networks” everywhere (whatever that mathematically nonsensical phrase is supposed to mean) are perversely choosing to ignore the reality of what is happening. It isn’t the 1% fat cats who are becoming victims of the new economy. It is the little capitalist in the middle. The brave little Jeffersonian middle class capitalist who could.
The long tail and the big head together are killing the middle everywhere. In education, locust swarms mean that this is a very bad time to be a mid-level education provider. If you’re not a top professor at a top university, or a long-tail independent provider of some sort of personalized education (such as bloggers offering courses on a very small scale — a game I am trying to get into), you’re in deep trouble. The education-seeking swarms will get everything they need from the top and bottom of the distribution and leave the middle to starve.
Hotels, taxis, education, music, publishing, restaurants. The list of locust-devastated/soon-to-be-devastated industries is growing longer every day. I suspect the locust economy is now bigger than California’s economy.
The smart Big Money is moving rapidly into swarming infrastructure (with apologies to John Hagel, I am starting to think “Pull Platform” is also lipstick on a pig — the 1% figuring out a way to use the 90% to feed on the 9% via a temporary alliance for mutual benefit).
The smart Little Money is moving into locust makeovers. People with little capital are figuring out how to arm themselves to be better locusts.
Is there a way to reclaim the middle? Is there hope for the Jeffersonian Middle Class?
The Myth of Jeffersonian Resilience
The Jeffersonian middle is the place where decency, goodness, hard work, an admirable work ethic, modest and non-predatory entrepreneurial ambitions and mainstream culture live. This is not the regular middle class but the Jeffersonian middle class that idealistically seeks an autonomous, meaningful existence.
The typical member of this class does not want to either build a billion dollar company or live off a passive-income lifestyle business based on exploitative arbitrage. He or she wants to work hard at a meaningful activity and asks only to be left alone to be content with modest rewards and economic autonomy.
These are the Clueless of the economy.
In fact both zombie and vampire narratives (and this locust narrative that I hope catches on) are based on the fears of this class. Even though the class itself is relatively small, it also serves as the class that the rest of society aspires to, or pretends to aspire to.
Wage slave and Big Company Executive alike dream of retiring to run small bed-and-breakfast or restaurant operations. So all like zombie and vampire stories.
It is also the class that foolishly believes in resilience without mobility at a small scale. Unless your idea of small-scale resilience is totally isolated communes behind an economic firewall that seals out the global economy (I have yet to see a working example of that ideal), you are coupled to the larger economy and vulnerable to bigger actors — Hamiltonian swarming-platform actors who believe in growing as big as they can, as fast as they can.
Because you see, it is not the Big Guys alone that you have to figure out. It is the Big Guys in an alliance with locusts whom you are predisposed to be contemptuous of, but are capable of sneaking through your defenses.
Being in the locust swarm full-time or part-time, sucks. But it is often the only way to make ends meet in the long tail. Being in the 1% (assuming you are okay moving from being slightly evil to Mr. Burns evil) is by definition not an option open to all.
But being in the Jeffersonian middle is probably the worst position to be in. Jeffersonian middle class dreams are not resilient even in grasshopper economies.
Unlike truly secure capitalists, who generally have enough wealth to cash in long positions in a declining economy for privileged starting positions in new ones as investors (Rockefeller’s descendants are still pretty darn rich several generations later), small-scale capitalists barely have enough resilience to ride out the economic volatility of a single human lifetime.
To be totally brutal about it, if you’re in the middle, you’re probably screwed no matter what the economic conditions are. Since industrialization began, there has rarely been a period where small business has been good business. If you’re not willing and able to grow big, your fate is to get eaten alive by either locusts (or directly by bigger organisms) or go out of business.
The resilience of the Jeffersonian middle class is a myth based on the false sense of security that comes with owning a business rather than relying on a paycheck. A very special kind of self-sustaining myth based on continuous churn.
To take coffee shops as an example, an unending supply of idealistic wannabe cafe owners enters the sector every year, operates at a loss for a few years, and exits. The result is that even under normal business conditions, without swarming locust consumers, this is a loss-making business with an extinction rate of around 90% at the 5 year point in the US. Starbucks has the scale to be profitable and resilient. Locust coffee drinkers happily drink the excellent, loss-making coffee from small, local Jeffersonian coffee shops and callously retreat to Starbucks or DIY homebrew if the prices go up.
Starbucks survives, coffee drinking grasshoppers survive, small coffee shops go in and out of business.
Add the locust-swarming platforms into the mix and you get mechanized, efficient predatory dynamics that speed up the idealism-to-extinction churn rate.
Throw in a recession, and instead of the devastation of true abundance (such as a harvest season), you get the devastation of a system in a state of scarcity that is trying desperately to send fake abundance marketing signals, in a hopeless fake-it-till-you-make-it attempt at survival.
The Fate of the Locust
Being a locust, and allying with the Big Guys to prey on the Medium-Sized Guys is no picnic either. Ultimately, it is a pact with the devil. Locusts are created out of a downwardly mobile paycheck middle class via dynamics I’ve written about before. Becoming a locust is smarter than staying trapped in either a doomed paycheck lifestyle or a faux-resilient Jeffersonian middle class. But only by a whisker.
The blogging/publishing component of my business (which is much smaller than my consulting business) is primarily an intersection of many locust economies. I am a prosumer locust in an alliance with Amazon, Google and various other locust-swarm pull platforms.
But what happens to locusts when all the crops have been eaten? To take an example close to home for me, what happens when traditional media is really gone, and there’s nobody slogging away to provide blogging fodder for locust media types in armchairs, like me? Almost all of my own consumption is traditional media — books, news stories and such — that I am helping Amazon and Google kill. I don’t really read other bloggers much.
Turns out, in nature, locusts most likely turn to cannibalism as a way to manage both supply volatility and secular supply decline during their swarming phase.
But ultimately, the swarms disappear. The intent of the swarming phase — riding temporary abundance to survival of the species — has been accomplished. In some cases, as with the Rocky Mountain locusts, if other factors come into play, they may even go extinct. In this particular case, the likely explanation is that farming destroyed vast quantities of locust eggs through large scale plowing of Midwestern lands via industrialized agriculture. So the small farmer and locust both disappeared, giving way to large scale agribusiness.
If traditional media gets killed and bloggers are forced to primarily work off each other’s work, an unpleasant endgame and shakeout will be triggered. I hope it does not come to that.
You can observe such cannibalization dynamics in the internal, so-called p2p dynamics in any locust market. Truth be told, participating as a provider in a p2p network like Airbnb or sharing a car is simply more inconvenient than owning your own place or car and having it always available.
When software eats everything by turning inefficient grasshopper consumers into efficiently swarming locust consumers, we pay for small new income streams and cheap deals by way of increased economic coordination costs and shadow labor.
In other words, in a locust economy, you cannot just decide to go somewhere and get in your car to drive there. You have to coordinate with other potential users of that shared resource. You have to keep your apartment clean and sharing-ready. You have to do minimum-wage work that you might consider beneath you (though such status concerns don’t bother me, annoying chores do).
In the sharing economy, we may not be eating each other literally, but we’re certainly eating into what Richard Dawkins called the extended phenotype of our neighbors. To the extent that your belongings are a logical expression of your genes and memes sharing them amounts to allowing others to eat them.
So the harsh bottomline of the locust economies, once the Jeffersonian middle class prey base has been bankrupted, is that we locusts turn on each other.
We call it peer production and prosumer economics, but it isn’t Jeffersonian producerism. It is locusts in their cannibalistic phase.
When the harvest is gone, software eating everything translates to prosumers eating each other.
There ain’t no such thing as a 50%-off lunch.
Now if you’ll excuse me, I have an expiring Groupon to use across town, and a meeting with a fellow locust to get to after. I hope that Zipcar downstairs is available. And then I have to get back home and clean my apartment — great use of a PhD in aerospace engineering, eh? — before my Airbnb guest gets here.
(Okay, I made up most of that last paragraph. Only the meeting-another-locust part is true. For now.)
About Venkatesh Rao
An interesting angle on a topic I’ve been thinking a lot about lately.
The share economy (aka cutting costs by cutting out costly human labor and binary property rights) is disruptive but I don’t think it’s inherently bad. It’s only bad because it’s breaking our systems for moving value around the economy… ok–and recreating the economic system on a giant corporate scale where there is no agency for smaller-scale (“Jeffersonian”) actors. So maybe I agree with you a bit.
I’m still not comfortable with the locust metaphor though. Basically you are arguing that we are short-term selfish utility maximizers at the expense of longer-term societal value. But is it really a tradeoff? I’m not sure we have to have either car shares or taxis and mass transit–much of this sharing may have marginal impact. (groupon is only the nail in the coffin anyway) And in industries where it does wholly cut out the middleman (say, when self-driving cars take over) the problem is not, per say, that we don’t have a person in the driver’s seat. The problem is that we have destroyed that particular way for a recent immigrant to earn a decent living for his family. I think we have to differentiate that because it’s a real possibility that sharing (and technology) can really improve things.
A further question I have is, is the tradeoff (between locust-like destructive consumption and consumption that fosters long-term value) really made at the level of each individual consumer? I don’t have much faith in the ability of the buy local/buy small movement to succeed in the face of continued layoffs and decreasing real wages–I think to a large degree its a matter of technology and to another unknown degree it’s a matter of policy. Not that massive income redistribution or redirecting antitrust policy toward producers (read Barry Lynn’s Cornered if you haven’t already) are easy solutions either. Probably we need to rethink corporate ownership and our financial system to really get any kind of humane system.
I understand (and agree with) the locust analogy as it applies to things like GroupOn: GroupOn encourages businesses to engage temporarily in self-canibalizing behavior in hopes of gaining future advantage. It’s just the old concept of the loss leader applied at scale (the social aspect), and with fewer controls to avoid catastrophe. The distinguishing feature of locust behavior is that it uses up the resources in question with no thought towards replenishment or regeneration.
But I don’t see how the analogy fits something like Zipcar. True, you are sacrificing some convenience for the sake of saving money. If you drive a lot, it’s entirely possible that you will wind spending more on Zipcar than you would on your own car, but that’s just poor personal economic decisionmaking.
The effect of car sharing is that the car, which would otherwise sit idle and slowly depreciate, gets used a much higher percentage of the time. It thus also falls apart more quickly, i.e. is “used up” more quickly. But the business model of these car sharing companies includes recovery of the cost of replacement of the cars, so “use it up and then abandon it” is not a feature of this system, unless the car sharing company is spectacularly poorly managed.
I guess I can see how this is locust behavior from the point of view of the auto manufacturers, to the extent that their business model requires selling cars which are mostly going to sit idle most of the time (and thus selling many more cars). But I don’t get the impression that was what you meant.
Caveat: I see this applying to car sharing to the extent that it applies to our entire economy – which is a significant extent. The reality is that our entire economic endeavor is a locust system, because it relies on using up a finite resource without worrying about replacement costs or long-term viability. Or perhaps it is merely predatory but not of a “locust” nature since it is, by definition, global while your locust analogy relies on local temporary surpluses in a landscape of scarcity. (Although, by that definition, the industrial economy is exactly a locust swarm if viewed temporally, in the context of human history.) Whatever: What I mean is that I don’t see how the locust analogy applies to car sharing any more than it does to the rest of our economy.
Think of it this way. If per capita car ownership falls from 1 to 0.5 in a community, and we think of car+adult as an extended phenotype human, the group just ate half its cars in a cannibalistic way.
The predation victim would be cab drivers.
So all forms of resource efficiency are locust behaviors? That seems to be what you are saying.
The predation victim would be cab drivers.
Also, this isn’t really true, based on my experiences as a CityCarShare member in Oakland. I use both that and taxis, for entirely different sorts of trips. CCS lets us maintain a 1 car household; it doesn’t replace taxis. Maybe in more dense urban areas (NYC) where car ownership of any sort is a huge PITA then it would be different.
_The predation victim would be cab drivers._
This has been my experience with Lyft in SF. I sold my vehicle, no longer take cabs, and have a much more pleasant riding experience getting to know the Lyft driver. Let the Locusts eat cabs… Lyft drivers make good money and seem happy to have a flexible work schedule and ownership over their vehicle, two luxuries most cabbies do not have.
Zombies are zombies, whether they think they are smart or self-interested. Still slaves to cravings.
Nice piece for thinking on. I don’t think you can conflate the GroupOn Ponzi scheme with the sharing economy, though – even if you ignore the Ponzi aspects of that particular endeavour. Let’s take the AirBnb example, and housecleaning.
Let’s say you have an extra room, and you rent it for two nights, once a week. You can pay for the entire apartment cleaning for less than that – and that generates additional income for someone. Given the AirBnb demographics, you’ve more likely taken that rent away from a large chain than a local mom-&-pop motel, while transferring it to a local relatively-poor business. Odds are that money then circulates more locally and more rapidly, increasing local money supply and velocity, while monetizing a potential non-depleting asset. The additional perceived value for the renter may be psychological or social – either way, price is only part of the value received.
Additionally, due to local regulations (occupancy, for example, or allowed parking spaces) actual sharing economy enterprises are forced to deal with local governments, which means *ahem* local social entrepreneurs building relationships with their local governance…. I think I read something about that in a Forbes blog post…. :)
Couponing is a whole different kettle – the creation of a swarm is indeed the purpose. Given that we know price is not the major consideration for most consumers (or is one of many), discounting as a tactic attracts the swarm while cutting the margin on goods with variable expenses. It’s no different than doing your after-Christmas sale on December 24th (which was killing margins a few years ago as stores did that.) Couponing simply reduces the money flow, rather than redirecting it.
So two different socio-economic systems are working here, with very different value propositions.
Also, a disclaimer: I participate in the Bay Area Sharing Economy Coalition, along with many other for-profit and non-profit enterprises, including both hospitality and transportation firms.
Airbnb is more in the cannibalistic rather than predatory phase of locust swarming, as I argued later in the post, which you may have missed. Through more efficient utilization, a collective lowers its total footprint. Prosumer = extended phenotype cannibalism may be a bit of an overwrought analogy, but I think it hangs together.
Has AirBnB ever gone through a predatory phase? I can’t really see that. If it doesn’t go through the predatory-cannibalism-crash cycle, then is that significant somehow?
Well…. if you look at the SOP of Airbnb hosts, they start with a low price to build up a rating and then jack up the price. Many people surf the low-priced host couches. I’ve done that myself… look for new hosts who have few ratings/stays completed to get a deal.
But it’s not as dramatic as with other sectors, since people are more picky about where they camp out for a night than where they eat.
“what we all understand in the abstract”
“operations of known dubious repute like Groupon”
You’re the last person I expected to slam something without having a nuanced understanding of it, and instead repeat the current media story. I could talk for an hour about Groupon, if you’re ever in the Bay Area and want to catch up.
I also think that using the 1% in this analogy confuses your point. None of the locust-generating businesses were started by rich people, they were started by normal people hungry for growth and opportunity. They’re the ones that recognized an abundant resource (value not provided/captured by Jeffersonian middle/underutilized assets/etc) and unleashed the mechanism to harvest it. It’s more like some grasshoppers enabling and unleashing the locust swarm – leave the 99% and the 1% out of it. Or maybe say that when locusts are enabled, it’s the 1% that can protect themselves and the other 9% that are worth harvesting are too weak to withstand the swarm.
If you really wanted to claim that the 1% enabled the 90% to eat the 9%, you would have been much better sticking to the finance industry rather than the tech industry.
A nitpick, Zipcar isn’t part of the sharing economy. It’s just a traditional rental car company that parks its cars in odd places and has a membership fee model. Getaround would be a better target, though less alphabetically satisfying.
The UX metaphor is definitely shared access, not renting in the traditional sense. The accessibility and short-term nature make it a sharing substitute for an owned vehicle. Traditional rental by contrasts extends the “ownership” experience to travel.
It is true that their marketing strongly sells the shared access concept. But anyone who uses Zipcar (myself included), very quickly realises it’s just another form of car rental: pay-per-hour and more convenient pick-up locations, but car rental nonetheless.
You make a number of interesting points in this essay, but I too am struggling to understand how Zipcar is part of the locust economy, unless car manufacturers can be considered Jeffersonian capitalists…
Great! The economy of early 21th century in a nutshell!
There’s been plenty of analysis, books and lecture tours about the Long Tail and the Short Head. For some balance, it’s time to do some analysis of the Fat Middle. And so, interesting to see your comments above about the 90% and 1% eating the 9%.
I think of the Fat Middle as also being the Volatile Middle. It’s where all the action happens as new stars emerge out of the darkness of the long tail, and where mega failures end up if they’re not snapped up. There’s a constant volatility as actors become big enough to be interesting but not yet too big to fail.
Yes, that’s exactly right. It is fundamentally a volatile place, an unstable equilibrium between two stable ones that is easily disturbed.
Venkat,
Great post. Lots of stuff to think about.
Locusts, cancer cells, and human psychopaths all share a common thread: local (individual or species) fitness at the expense of an ecosystem. Jury’s still out on whether we’re the best thing that ever happened to the universe (as some religions claim) or a cancer species in the ecosystem (as pessimists argue). Ultimately, though, the struggle for survival leaves no room for good and evil; those are luxuries of the comfortable. In nature, it’s *fear* (and various tricks that biology gives us to turn those impulses into something more pleasant, e.g. thrill and sexuality) that drives most action. Indeed, most locust behavior comes out of fear of non-survival. They’re insects, so they don’t tear ecosystems apart to “be evil”; they do it in response to natural stresses.
Agricultural people were the first locusts– much worse for the environment, able to create slave economies and generationally-persistent hierarchies, tending to treat women as chattel in huge harems, and prone to poor individual fitness due to malnutrition– but 10+ times as populous, per unit land, as hunter-gatherers. They weren’t as quick in their movement as locusts, moving over generations rather than in days, but they had a similar effect. Over about 110 centuries, they turned the semi-arid but extremely fertile (animal traffic) reaches of Northern Africa and the Middle East into a harsh, almost uninhabitable, desert by agricultural people not wanting to ruin the land, but just trying to survive.
From -200,000 to about 1850, there practically was no improvement in the human quality of life. There was (very slow, until 1700) economic growth, but population gobbled it up. Malthus was only “wrong” insofar as the industrial revolution brought economic growth into the 1-5%/year range, instead of the industrial era’s 0.1%/year range, and birth control (starting in the mid-20th century) put the brakes on population explosions. Malthus was dead wrong about economic growth being linear (in 1798, it was a slow exponential, around 0.5-1.0% per year, that *looked* linear) but his conclusions (“catastrophe”) were accurate. England didn’t have a Malthusian catastrophe in the mid-19th century, but one happened; they just outsourced it to Ireland. (Yeah, we still remember.) Europe stayed healthy by outsourcing its Malthusian problems via colonialism until industry reached a point (1960? birth control is a major player here) that the Malthusian issue became no longer a threat, except in the poorest corners of the world.
I’ve summed up my long and expansive Gervais / MacLeod series in the 21st part, here: http://michaelochurch.wordpress.com/2013/04/03/gervais-macleod-21-why-does-work-suck/ . I focus on Convexity, which is the most important (in my mind) economic issue of our time. I think you’re basically right that the terminal middle class of the Jeffersonian ideal is a dead end. It has been for some time.
In 1800, we were still in the agrarian era. Economic growth was slow– maybe 1% per year, if you were lucky– so one *had* to find a way that people could be content with economic mediocrity. Not everyone could be rich; each rich man made 100 others poor (or unfree). That’s why religion was so important in the agrarian era: it gave people a plan that enabled them to live in a slow-progress, practically zero-sum, world without killing each other, populating too fast, or starving themselves to death. Jefferson wanted to replace religion (which he recognized as mostly nonsense) with something that could be equally stabilizing, but treat people as rational adults rather than children being told bedtime stories. He had to make the yeoman farmer an ideal *without* appealing to supernatural authority, so he used the classics and Age of Reason idealism instead.
The agrarian era wound down over the 19th century, and industrial era (1800 – 20??) brought us visible economic growth that we now take for granted. If you’re rich, you can park your money in the growth process and be damn sure you’ll see *some* interest on it. You don’t have to make anyone else poor to do that. Industrial operation also has the neat property of providing concave labor (low performance risk; difference between mediocrity and excellence is small) to the masses. Jefferson’s yeoman farmer ideal never worked, but those people could move to the factories and participate in the industrial process in easy, concave labor that left them with very little income risk due to the hourly wage.
Work tends to fall into one of two categories. Concave work means that the input/output curve– the relationship between factors like talent, skill, effort, morale, and wage level; and productivity– is concave, so the difference between noncompliance and mediocrity is large but that between mediocrity and excellence is insubstantial. It’s low-risk, mundane work where mediocrity is acceptable. Concave work built the large middle class of 1925-2000. Convex labor’s the opposite: the difference between mediocrity and excellence is huge (“10x programmer”) and that between mediocrity and noncompliance is small. It means you probably shouldn’t show up to work if you’re going to be mediocre; you probably won’t get a job in the first place. The implications for job security, education, and employment are *vast* and I really think that Convexity is *the* economic problem of the 21st century. It’s *why* we will never again see a society where everyone between 18 and 65 can get paid work at a living wage. (We’ll need to implement basic income at some point, but I wouldn’t hold my breath, given the US political climate.)
Software isn’t eating “the world”. It’s eating the concave commodity work that just happened to be 90+ percent of paid wage work in the industrial era, and 99+ percent of the work for where there’s *regular* pay. (Convex work is highly sensitive to small differences in performance that the concave world could ignore.) It comes down to this. Convex labor is work where the “saturation point” is so far out that no one has found it yet. It’s so far away to the right that the logistic (“S-curve” in what they call it in Econ 1o1 to avoid scaring future investment bankers with math, but that’s what it is– a logistic) looks like an exponential curve. The work is very hard, but if you do it well, the rewards are extreme, and no one knows yet what the upper limit is. With concave work, we know what perfect completion is. If we know perfect completion, we can specify it. If we can specify it, we’re either able to program it, or it’s a field of active machine learning research. Software isn’t eating “the World”. There will be plenty of World left. It’s just eating all of the concave labor on which the risk-intolerant middle class (who fall into poverty if a few paychecks fail) relies. We can’t compete with machines. We thought *cab driving* (concave labor that required humans until recently) would always be done by people. We’d have said the same thing about sorting mail (optical character recognition) in 1985. Yeah, about that…
The Locust Effect is seen in the MacLeod Loser tier of the broader society, which is probably 90% or more, and largely driven by fear and insecurity. As you said, no one wants to put restaurants out of business when using a Groupon. Nor do Wal-Mart consumers want to destroy small-town economies, brutalize illegal immigrant workers, or support factory-owning thugs in China. I grew up in blue-collar Pennsylvania; no, these people aren’t dicks. These things happen because people are unaware or indifferent. As Losers, they’re subordinate (they accept their mediocre lot in life) and strategic (they make sound economic choices) but not dedicated (they don’t *care* enough to fight a broken system).
The Clueless haven’t figured out that Convexity is a permanent affair. There *is* convex service work, but you have to be creative and energetic to do it well enough to get paid and stay afloat. If you’re continually coming up with new reasons (or a reliable but important old one, like excellent food at a reasonable price) for people to eat at your restaurant (convex creativity) then you might stay in the game. Convex labor’s much riskier, so there are no guarantees, but you have a chance. However, if you persist in doing the concave stuff that worked for the past 200 years, you end up in a race to the bottom. MacDonalds can win that race– they’re doing concave/commodity work, but have the resources to win sieges– and the little guy can’t. He loses.
That’s the big problem with Groupon. It’s a concave/cost-cutting mentality. It’s not helping people increase value by finding the really interesting (“convexly” successful) restaurants but cut costs.
I also think Groupon’s getting bad PR because it’s in Chicago and therefore can’t be a VC-istan investor/press darling. I don’t think that what they’re doing (making money in a way that has nasty externalities) is any worse than what most of these “social media” outfits are up to. But that’s another rant…
Your last paragraph makes some pretty strange assumptions. Being in Chicago didn’t stop “VC-istan investors” from investing in Groupon – it raised $1.2bln from Andreessen, Battery, KPCB, Accel, SV Angel, et al. The tech press was effusive with praise in its early days, and if anything, persisted in running mythical stories about Andrew Mason as embattled hero founder rather than inept public-company CEO running his company into the ground.
Coverage detailing a ~75% drop in value since the IPO is not “getting bad PR,” it’s just non-bullshit reporting.
Whoa, this is an essay in its own right. Lots of thoughts in there. No immediate response, but noted and simmering.
I don’t like convexity as a broad lens thought… Taleb uses it too. Like “Bayesian” it is too technical a metaphor for my taste. It obscures more fundamental qualitative things going on I think.
I’ve been sort of sampling your extended series off and on, but I think you’re chasing down different bunny trails than I am personally.
The reason I use “convexity” is because it’s emotionally neutral. Sometimes, one hears phrases like “creative economy” (cheerleading, positive) or “star system” (negative, if you’re not a star) or “power law” (which has the same problem of being over-technical). Those aren’t as core to the problem, anyway.
I also prefer “convexity” over this Singularity nonsense because there’s just no evidence that machines will be ready to take over *all* human work in any timeframe I will live long enough to care about, but the concave work is obviously theirs. Singularity might happen in 2030 or 2175 or 4000 or never. Convexity is happening right now.
I use “convexity” for balance. Yes, there’s a lot of good in a world where commodity work is all done by machines. However, this is a massive social instability problem if it’s not managed well. I’m trying to avoid using words for it that have emotional baggage attached. It’s not good or bad. It just is. It’s the economic reality we are (probably) going to face in the 21st century, and we’re probably all a bit out of our depth.
The word “convexity” is perversely technical. I don’t think there’s any way to describe it to the average person except to draw a curve and hope he gets the picture instead of saying, “Huh?” Perhaps “risk-intrinsic” is slightly better. Concave work can provide value without risk, although that’s usually to compensate for its unpleasantness– which is why someone out there is working on automating it– while convex work provides value *because* of its risk and creative intermittency.
I see what you’re saying about vague use of the word “convexity”, but isn’t the word perfectly useful/acceptable when there’s a generally understood numeraire?
Of course, that was one of the few unavoidable gaps in Antifragile–“good” and “better” are numeraires you have to take on some faith (works reasonably well for strength gains or financial gains, but harder to do for boosts of things that are more complex.)
What is the predation victim for AirBnB? That locust economy seems like it makes it easier for apartment renters/landowners to keep their physical property. Hotels? That doesn’t seem like a bad long-term problem for the locusts.
Airbnb is an interesting tiered system. There’s pure consumers, prosumers and pure hosts within the p2p network, and the coupled substitute market of hotel rooms.
The devil’s in the details on this one. I saw an SF map which showed that Airbnb properties were often in areas not well covered by hotels, so the direct competition rate may be low.
Still, there’s got to be some net loss for the hotel sector, and in more suburban areas, that means small biz owners (motel franchisees). It’s not direct predation though, it’s death by starvation for them. You can imagine them trying desperate Hotels.com discounting deals to compete if Airbnb type things threaten them too much in some markets.
I think you’re painting daily deals and collab consumption with the same broad brush, and I’m having a bit of difficulty with that aspect of this piece. The equilibrium problem is potentially a killer for both of these models; at a minimum, I think it means that most of the venture-backed aggregator businesses to grow from the “sharing economy” are going to fail. I don’t believe that the margins of a p2p economy allow room for a middleman cut. If they reach the kind of large scale they need to support themselves, they will be sitting on a market that will feel borderline predatory for most people on the sell side.
BUT, that said, I do believe that Airbnb is an exception to this…it actually does have solid economic value for both sides, and quite often fulfills the idealist “sharing economy” vision of a more personal, unique, and enjoyable experience. Perhaps at scale that falls away?
Anyway, this piece was inspiring (literally: I was inspired to blog a response). So, thank you for your work.
I would be interested to hear how the locust/swarm behavior analogy could be developed to cover behaviors around fashion. Especially within the context of scarcity/abundance but taking it to a scarcity-as-exclusivity and abundance-as-non-exclusivity angle. In other words I wonder how this could apply to not just bargain hunting but cool hunting as well, or even cool bargain hunting? I wonder if similar pieces of the brain light up in either case?
Visuals: the opening of a new store from a popular chain or launch of a new product from the same, with a line of shoppers outside (such as the Nike store in downtown Seattle) or a swarm of ‘pickers’ scouring local thrift stores on days when they get a dump of new inventory, looking to score scarce high value items that a Jeffersonian boutique might take (such as Pretty Parlor on Capitol Hill). In fashion, whether produced and sold by a big supplier or secondarily merchandised by a Jeffersonian boutique, part of the success formula is to finely modulate between scarcity and abundance. Presumably, a $400 sweater from a popular label on sale for $200 would find a happy and willing locust buyer. But if there were 400 copies of that same sweater at that same price in that same retail location (abundance) and that same locust saw 200 other locusts in line with the identical item clutched under wing maybe not so much. That saturation or satiation deflection point causes another cycle or trend trigger, and off these locusts go, looking for the newest crop of fashions.
The fascinating use case for me isn’t admittedly as important as the economy-level stuff such as declines of certain classes, but it could be fun to locate predictive aids in all this that would help in guessing where this non-migratory but truly randombulatory swarm might go next…
Hmm, also reminds me a bit about patent trolls: http://www.rackspace.com/blog/why-rackspace-sued-the-most-notorious-patent-troll-in-america/
“Meanwhile, IP Nav wears its designation as a patent assertion entity, or patent troll, as a badge of honor. It makes the laughable claim to be a “white hat” patent troll created to “give the little guy a chance.” This is tragic comedy at best. There are few trolls more notorious than IP Nav, and there is no such thing as a patent troll that has the best interests of small businesses in mind. Instead, IP Nav and Parallel Iron are acting in their own selfish interests and suffocating innovation, while stripping capital away from businesses both large and small. Everybody knows they are a duck – or should we say, a troll.”
I’d just like to leave this interesting observation…
Locusts are an excellent, easily gatherable, source of protein.
They often provide -better- nutrition during a swarming event than the crops they displace… IF you are capable and willing to change your diet. Several more “primitive” groups welcome the locust swarms for their bounty, while most “industrialized” groups bemoan their crops and then either taint the swarm with poison or let it fly away.
I’m not really sure how that properly slots into the analogy, but it’s worth a mention.
Hehe. Entertaining thought.
I think the analogy is to properly tax Groupon, Amazon, Google and other big players at the point of sale and not Ireland.
I really, really like this idea that doesn’t quite knit together for me. My brain parts light up with the comparison of people in social networks and locusts. Often networks just allow people to fall for things faster. I love the exposure of the self-righteous, altruistic illusion that social networks often drape themselves in.
But I’m not buying the weakness of what you call the Jeffersonian Middle Class. Taleb’s ideas of Antifragility make more sense to me.
Large organizations don’t respond well to change. Highly specialized people do not respond well to change. And the more that they use what power they have to insulate themselves to change, the less prepared they are for the inevitable swings of history.
Which explains why we’re not run by the Borgias. Or Caesars. Or the descendants of Charlemagne.
But I’m loving the analogy and appreciate your thinking. Twitter is swarm behavior for sure.
Very cool piece Venkat, love that you’re willing to extend these ideas!
A couple of questions though… how might uncertainty (as opposed to Risk, which is at the relatively certain end of the spectrum… ie see insurance) play into this metaphor…?
Also, curious how you might see phenomena such as Crowd Capital (see here: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2193115 ) from both the organizational and “locust” level?
If you disapprove of it, you can chose not to participate in locust behavior.
I really enjoyed this piece. You might consider adding it, or some form of it, to your Gervais principle series. By the way, any news on when the latest part in that series will be posted?
To quote the article: “It is no accident that the worst-hit victims of the locust plagues of the 19th century were small farmers living the Jeffersonian dream handed to them by the Homestead Act of 1862.”
Were the locusts acting intentionally against small farmers or was the Homestead Act designed by Jeffersonians to feed the locust swarms? I’m not clear on which part of this wasn’t accidental.
Short version: They had all their few eggs in one small basket. Richer people are more diversified, poorer ones are more mobile/adaptable.
The way I see it, the Jeffersonian middle class is the only class that consistently had a large quantity of debit relative to their assets.
The poor tend to only have transient debt to make ends meet, and no one is willing to give them more. The rich tend to have a lot of dispersed assents or at very least a good bankruptcy lawyer. The super rich can be too big to fail.
But the Jeffersonian middle, the small farmer or business owner, can be significantly in debit with their only collateral being their main asset. This means that they are very weak to black-swan events that can greatly devalue their asset for a short period of time. The 2008 housing crash is an obvious example.
That’s an interesting idea, you could say they are the lowest income group equiped to make large long term bets and then bank on the returns, (indeed will be required to do so by various fixed costs associated with their business), but not high enough on the income scale to hedge those bets effectively.
This compares to lower income gambling, which is almost universally considered a flat cost or hobby, (and this even includes risky capital investments) to be tolerated at the side of a consistent up front income supply.
Really really interesting…and honestly really cool horror movie analogies. We just wrote a post about this:
http://startupbook.co/2013/04/09/are-you-a-locust-or-a-zombie-how-groupon-is-the-horror-story-of-our-new-digital-economy/
Definitely going to be reading ribbon farm more often.
Torrenting is a good example of locust-like behavior. A huge swarm of people take the artistic product of musicians, actors and writers, while at the same time failing to contribute any value to those industries, and thus endangering their very existence.
nice.
Excellent essay. After reading Mr Jefferson’s Lost Cause by Roger Kennedy, it’s hard not to read Jeffersonian Democracy as an ironic phrase, considering how much Jefferson did to further the cause of slavery.
The Crusades were a locust swarm.
I’m not sure about the net impact of “shadow labor.” Yes, we’re expected to do things to replace travel agents and gas station attendants. But we’ve traded that off for all sorts of home maintenance, food prep, and cleaning productivity that are no longer being done by consumers.
We used to buy raw ingredients for meals, along with simple cleaning supplies like bleach, ammonia, and brooms. We input a little bit of home productivity in cleaning our dishes and mops, assembling our meals, etc. That has fallen by the wayside (see “Forks over Knives”), turning the average home consumer into a purer consumer who purchases lower and lower entropy good every year, reducing that existing shadow labor.
So now we do “shadow labor” outside the home during the non-fungible time we would have waited on a gas station attendant, or from the home to replace inefficient middlemen like the travel agent. What’s the net, in terms of hours per day doing non-wage productivity/entropy reduction? How many non-wage hours are spent being purely consumptive? I suspect we spend less time being productive outside our jobs.
Admiring the time and energy you put into your blog and in depth information you present. It’s awesome to come across a blog every once in a while that isn’t the same unwanted rehashed information. Excellent read! I’ve saved your site and I’m adding your RSS feeds to my Google account.
f3.toThis is good stuff. I’ll have to read it again when it’s not so late at night :)
Dumbest segue into an article about locusts, ever.