This is essay 2 of 4 for the BanklessDAO Writers Cohort.
I think I have opened a can of worms, or perhaps a can of cobras, by choosing this topic for the Writers Cohort task this week. My planned focus for this piece was examination of the factors which can influence the success or failure of incentives, so that these can be taken into account when designing incentive programs for DAOs. However, the reading I have done in preparation suggests that this is a huge topic, with multiple strands, so I may just take a peak and put the lid back on for now.
An incentive, in my view, is a reward, identified beforehand, to encourage a certain behaviour. I intentionally held off searching for a definition of incentive before I wrote that sentence, so let’s see now whether I came close.
The Cambridge Dictionary Online defines incentive as “something that encourages a person to do something”. The Oxford Learner Dictionary has a similar definition, noting that incentive could be both “countable or uncountable”, and also gives the origin of the word as “late Middle English: from Latin incentivum ‘something that sets the tune or incites’, from incantare ‘to chant or charm’.” Was that what I meant by cobras? No, keep reading.
The dictionary definition of incentive might be fairly straightforward, but designing an effective incentive, something that will elicit the desired behaviour, can be complex. That’s because humans are complex creatures. While it may be possible to come up with an incentive that appears to work as intended, the tricky part is to examine what the unintended effects of such an inducement might be. A perverse incentive is one that produces an undesirable, in fact negative, consequence, and the cobra effect is named for one anecdotal example of this, although there are many others.
The cobra effect describes an attempt by British officials in colonial India to reduce the local cobra population by committing to pay a reward for dead cobras. This worked well for a time, but eventually backfired when people began breeding cobras purely to acquire the bounty payment. Once this became evident, the reward was discontinued. The unintended negative outcome was that the breeders, having no use for them, then released all their cobras back into the wild, and this caused the cobra population to increase.
The attempt to simply identify a desired outcome (dead cobras) and then base the reward on that was a huge mistake. While dead snakes may have appeared to be serving the officials’ end goal — to reduce the wild Cobra population — they did not think through the fact that they would be creating a market for the snakes, which sellers would pursue without regard for the original intention.
Clearly, coming up with an effective incentive must involve careful consideration of the actual end goal and what the implications of each proposed pathway to that goal are. It strikes me that incentivising certain behaviours or steps along that pathway could be more effective than pinning the reward to the results, but I think the most important takeaway is that brainstorming, consultation, feedback loops, and holistic thinking are integral in the incentive design process.
For incentives that involve an element of competition, i.e. that the outcome for an individual depends on the actions of all the other involved individuals, I’m sure that there is also a need to understand game theory, but not today. There’s a whole body of literature on the interplay between incentives and intrinsic or extrinsic motivation too, which I think warrants attention, but here I’m focusing on the relevance of systems thinking to the process of incentive design. In particular, I’m inspired by the way this approach could contribute to the creation of continuous, positive loops: that ‘flywheel’ that you read so much about in Web3 material.
In their work to develop an agreed definition of systems thinking, the authors of this article provide a great overview of the concept. They refer to “an increasingly complex, globalized, system of systems future” and note that we must be able to identify and prepare for the “ripple effects” of our actions. I learned that two of the key ideas behind a systems approach are attention to the various parts of a system and the relationships between them, and the importance of feedback loops in understanding, connecting, and predicting the outcomes of different actions within the system. The article quotes Barry Richmond, the person who originally coined the term, as saying that “people embracing Systems Thinking position themselves such that they can see both the forest and the trees; one eye on each” (1994).
We can look to permaculture principles as an example of a holistic approach which emphasises care for the entire system and its constituent parts within the design process. According to the Permaculture Research Institute,
“The philosophy behind permaculture is one of working with, rather than against, nature; of protracted and thoughtful observation rather than protracted and thoughtless action; of looking at systems in all their functions, rather than asking only one yield of them; and allowing systems to demonstrate their own evolutions.”
Working in a DAO, it can sometimes be difficult to keep an eye on the forest; we tend to focus on our little stand of trees. The nature of a decentralised organisation is that small groups of individuals within the whole are often discussing, testing, implementing, or discarding incentivisation ideas without very much consideration of how they might influence the entire DAO’s productivity.
It’s worth regularly taking a step back to evaluate the growth and health of the organisation, and to ensure that there are mechanisms in place to care for the whole. In pursuit of that virtuous cycle, that flywheel effect, we should ensure that those mechanisms are designed to incentivise positive contribution and progress, to reward behaviours that create beneficial ripples, and to keep the snakes from our door.
I know that I have raised more questions than provided answers through this short piece of writing. One question I tried to find an answer for is whether a better solution was eventually found for the cobras. Although the anecdote is all over the web, the story neglects that particular aspect. I did find one site, called Unintended Consequences, on which author Paul Orlando lists some potential approaches to improving the efficacy of the bounty. Before you go and look, I challenge you to think about it for yourself — see what you can come up with!
