The reserve army of the unemployed ant?

Social insect colonies are highly successful, self-organized complex systems. Surprisingly however, most social insect colonies contain large numbers of highly inactive workers. Although this may seem inefficient, it may be that inactive workers actually contribute to colony function. Indeed, the most commonly proposed explanation for inactive workers is that they form a ‘reserve’ labor force that becomes active when needed, thus helping mitigate the effects of colony workload fluctuations or worker loss. Thus, it may be that inactive workers facilitate colony flexibility and resilience. However, this idea has not been empirically confirmed. Here we test whether colonies of Temnothorax rugatulus ants replace highly active (spending large proportions of time on specific tasks) or highly inactive (spending large proportions of time completely immobile) workers when they are experimentally removed. We show that colonies maintained pre-removal activity levels even after active workers were removed, and that previously inactive workers became active subsequent to the removal of active workers. Conversely, when inactive workers were removed, inactivity levels decreased and remained lower post-removal. Thus, colonies seem to have mechanisms for maintaining a certain number of active workers, but not a set number of inactive workers. The rapid replacement (within 1 week) of active workers suggests that the tasks they perform, mainly foraging and brood care, are necessary for colony function on short timescales. Conversely, the lack of replacement of inactive workers even 2 weeks after their removal suggests that any potential functions they have, including being a ‘reserve’, are less important, or auxiliary, and do not need immediate recovery. Thus, inactive workers act as a reserve labor force and may still play a role as food stores for the colony, but a role in facilitating colony-wide communication is unlikely. Our results are consistent with the often cited, but never yet empirically supported hypothesis that inactive workers act as a pool of ‘reserve’ labor that may allow colonies to quickly take advantage of novel resources and to mitigate worker loss.

That is by Daniel Charbonneau, Takao Sasaki, and Anna Dornhaus, file under “speculative.”  For the pointer I thank Eric Durbrow.


Communism? Minimum guaranteed income?

Yes, communism. The "Theory of the Firm" (Arrow, Chandler, others) posits that all corporations are communistic, since they get prices for many things 'internally' since the search cost to do so is actually cheaper than relying on market prices. So why isn't the whole economy communist? That's an advanced topic for another day.

Another finding from the above: corporate raiders who prune deadwood and get money with LBOs (trading equity for debt, then cutting fat to pay off the debt) are relying on a feature of all corporations articulated by Chandler and others: due to search costs, a corporation 'over-hires' for every position, so that if they need another candidate (because one dies or retires or leaves) they can do so quickly without the need for a costly search. So *all* corporations have fat. It's part of their DNA. Distinguish this from government offices? Arguably the same.

Bonus trivia: Edwin O. Wilson!

I agree.

Yes that was brilliant, you are almost high enough quality to be Brazilian.

I'm allowed to be a cuck because there's nothing expressly forbidding it in the constituuuution.

What I was going to say.

What he said

"Distinguish this from government offices?"

No one can LBO a federal department. When the fat never gets trimmed, it builds up. Then bad results happen, but nothing changes. Or worse, laws are passed to attempt to ameliorate the problem, but usually make things worse.

They also serve who only stand and wait.

Drat (or drant), you beat me to it.

We can also conjecture about ZMP ants. I have a vague recollection that a badly injured ant gets pushed out of the colony, or maybe it's bees that I'm thinking of.

ZMP ants get eaten by other ants. Among the most powerful deterrents to an ant invasion in its early stage is to pick up some ants, lightly crush them to death, and sprinkle the dead ant bodies on the trail. If you catch it just right, the ants will take away the ant bodies and then abandon the trail.

Sick bees die in the field away from the hive. I have a beehive in a hollow tree, and I often see worn out bees crawlling around that couldn't make it back to the hive.

'file under “speculative”'

Why? The experiment seems fairly straightforward, with empirical data which should be easily replicable.

Particularly as nobody is stupid enough to compare an ant colony to human society as speculation, one would reasonably assume.

The stated conclusion may be speculative. It might not be an explicit desire to keep workers in reserve, it may be a limit of management.

The complex state machine which forms the colony's "brain" may have limits, inefficiencies.

And of course that "brain" probably does not have "intent" in a cognitive sense.

I mean the conclusion may be correct, but we certainly can't interview an ant colony and ask about the slackers. So we risk over-conjuring a human explanation.

@prior_test3 - "particularly as nobody is stupid enough to compare an ant colony to human society as speculation, one would reasonably assume" - you are not familiar with E.O. Wilson, who makes exactly such a claim and he's no stupid fool.

@prior_test3- re-reading your comment, it's not clear what you mean...but my point is that E.O. Wilson did see parallels between human society and ant society, that's his one-sentence famous-for attribute.

"you shouldn't learn anything about our behavior from ants"

"Although this may seem inefficient....."

Yes. Because it is. That reserve labor force could be contributing to additional production that makes the colony larger thereby negating the "status quo" equilibrium restoration argument implied here. Making the colony larger means there's no status quo to go back to and likely also means losses are less severe, not more, when something bad happens. These ant loafers are not helping to grow the pie!

I think we know which strategy is the winning one in the long run in the natural world. Nassim Taleb seems to have crossed the line into crackpot-dom, but I can hear him shouting, correctly, "Fragile!".

The super-expansive colony that runs on all cylinders will outpace the other colonies ... until it doesn't. When whatever catastrophe occurs (anteaters, burrowing warthogs, termite invasion, whatever) that so-called efficient colony will have no reserves, more infrastructure than it can sustain, more fixed costs so to speak and not enough resources to maintain itself. So the so-called 100% efficient growth colony will do great for a season or two, and then disappear when things get tough. The surviving colonies will be the ones who give themselves a reserve, and thus are robust rather than fragile.

Econ 101 students learn that an economy that is on its Production Possibilities Curve has some unemployed workers. An economy with 100% employment is in an unsustainable disequilibrium. So is an ant colony.

However, you have a point when it comes to humans running businesses. A careful manager will plan to have a budget reserve, contingency plans, inventory to cover interruption, backup generators and data backup, etc. A super-expansive firm will not only run with reserves close to zero, but will often put itself into a potentially negative financial situation because it leverages and borrows to the hilt.

Most of those companies fail, but some will survive and a few of them will become the next Apple or Amazon. And when the financial disaster strikes, although they might disappear like the fragile ant colony, if they do have a superior product or technique or idea, they'll likely get bought out by some financially sturdy company or investor. So unlike the ant colonies, a modern market economy has mechanisms where the "DNA of the superior firms" can survive and be passed on even though the firm itself was fragile.

I agree with your argument about ant colonies - an ant colony might have a much higher probability of catestrophic events that wipe out large numbers of workers happening. If the workers spend their time enlarging the colony, all of that is wasted effort.

On the other hand, I'm not so sure about your idea that the DNA of the superior firm survives. IP might live on but the investors who paid for it still lose their investment, which would disincentivize future investment.

"On the other hand, I’m not so sure about your idea that the DNA of the superior firm survives. IP might live on but the investors who paid for it still lose their investment, which would disincentivize future investment."

All the investors that thought the firm that died was too risky and instead bought into the more conservative surviving firm reap the reward. So, future investment is incentivized to look for safe, but not too safe investment.

But then the ability of the investor to profit depends on the existence of riskier investments that the actual investor must eschew - so they can die and be gobbled up by the less risky firm. That doesn't make a lot of sense.

"That doesn’t make a lot of sense."

The good or lucky investor, picks a firm that is at the marginal point of successful risk. But know one can predict the future. So, you never know for sure where that edge lies. There's a reason nearly every bit of good stock advice starts with diversify your investments.


I too became indignant when I read of these loafers.

The appropriate perspective is to consider the ant hill, and not the individual workers, as the organism. Lots of organisms behave this way.

Any analogy to human beings or human institutions is very dubious.

Any analogy to human beings or human institutions is very dubious.

You wish!

Tyrone would say it is because the ants have really good computer games... ;)

Consider the crews, pilot, co-pilot and emts, of helicopters attached to the emergency rooms of hospitals. Those people are all being paid to sit around and watch Seinfeld re-runs until some doofus runs his Audi into a bridge abutment somewhere, maybe next week.

A Festivus for the rest of us!

I put the Festivus pole up my ass and was denounced by the Bishop.

Thus, inactive workers act as a reserve labor forceand may still play a role as food stores for the colony,

So they get eaten when the colony starts to go hungry?


I think that was a reference to trophallaxis, not cannibalism.

Insert your tongue in my butthole.

Eat my ass out!

Efficiency is applied only when needed. Survival is paramount.

The one week and two week time frame for replacement to what is considered an optimum.
The laying of eggs to adult ants is about 6 weeks.

So the quick replacement within a week is what? Another redundancy? They stop eating each other to replenish the numbers?

So ants never developed the ability to contract out maintenance labor and/or no other species evolved to provide gig workers for temporary fluctuations in ant colony maintenance needs. Perhaps because they never evolved the ability to trust others outside of their tribe/colony. Or perhaps because they never invented money. Or is the latter necessary for the former?

Incorrect! Some ants contract out labor:

That seems analogous to keeping dairy cows, and we generally would consider farmers as having "contracted out labor" to dairy cows.

This count?

"Formica sanguinea workers from established colonies are also capable of raiding worker pupae from nearby Serviformica nests. This raiding can provide food and new Serviformica workers. After eclosion these slaves behave as regular members of the society, performing normal worker tasks and taking care of the slave-maker brood. Since F. sanguinea is a facultative social parasite, the workers are capable of nursing broods also independently and slaves are not found in all colonies."

So contractors parasitize refineries! Never thought of it that way. Thanks!

More the opposite - the "slaves" of the Formica sanguinea have some similarities with contractors, in the sense that the Farmica sanguinea only occasionally captures slaves (unlike the Amazon ant, who need slaves to survive; then, for the Formica sanguinea, slaves are almost as workers recruited because there is a peak of production; "almost" because, afaik, they don't "release" - or even eat - the slaves when they don't need them).

I'm not sure that figuring out how to have aphids extract from otherwise inedible materials something that is then processed into ant-friendly honeydew is quite the same thing as solving the problem of surplus labor when demand fluctuates unpredictably.

The history of all hitherto existing ant colonies is the history of caste struggles.

Dear Ty-Dawg, why is it that we so casually draw inferences about human biology or behavior from observing rats and other critters, like pigeons for instance? I kind of understand observing primates for this purpose. But what does experimenting on rats tell us about humans? I'm sure there's an answer out there, since so many scientists are proponents of this practice. I'm just being lazy, and asking you, or any other readers on this blog. Thanks in advance.

Researchers just study the critters for their own sake, or they look for models which have wider overlap. Mice get cancer, so a mouse model may help you study cancer. Rats learn so they can help with a general learning model.

Rodents (or ants) provide much less overlap for unemployment in a market economy.

Comments for this post are closed