Some notes on conflicts of interest

From Bitcoin Debates
Jump to: navigation, search

Accusations of people being corrupted by conflicts of interest are frequent in Bitcoin related debates.

In the block size debate, those working at Blockstream are often accused of being motivated by a conflict of interest to advocate for small blocks, in order to make their services related to Lightning and sidechains more valuable. Those advocating for larger blocks (for instance Mike Hearn and Gavin Andresen, who are advisors to Circle and Coinbase, respectively) are often accused of being motivated by a conflict of interest because if Bitcoin gains wide adoption they'll profit off of all the extra users, and/or their personal Bitcoin holdings will increase in value.

What is a conflict of interest?

From wikipedia:

A conflict of interest (COI) is a situation in which a person or organization is involved in multiple interests (financial, emotional, or otherwise), one of which could possibly corrupt the motivation of the individual or organization.

Whether a conflict of interest exists depends entirely on the incentives of the situation that a person is in. It isn't about whether the person is actually corrupted or not.

Claims that a conflict of interest exists can't be countered by appealing to the person's good character

People often react very defensively when it is claimed that they have a conflict of interest. They often deny they have a conflict of interest based on the fact that they hold their views honestly. Given the definition above, this is irrelevant to whether there is a conflict of interest.

Humans are prone to bias and self deception

All humans are capable of bias. People can easily deceive themselves. They can believe something completely sincerely, and not realize the underlying reasons. It is usually not possible to introspect and determine whether your beliefs are arrived at via pure uncorrupted reason.

Some people are more difficult to corrupt with conflicts of interests than others, but you don't need to be a bad person or have weak character to be somewhat corrupted by a conflict of interest, because bias is such a universal phenomenon.

Corruption is a continuum, not a binary state. Someone may be able to reason in a situation containing a conflict of interest and be influenced only slightly by it, while someone else would let the conflict of interest affect them much more.

Existence of a conflict of interest does suggest that the accused is more likely to be corrupted

Although whether or not a person has a conflict of interest is based on only the situation and not the person, the claim carries the implication that a person in a conflict of interest is more likely to be corrupted than if they were not in a conflict of interest, based on what we know about human biases.

Having a conflict of interest doesn't mean your reasoning is definitely corrupted, it just makes it more likely that it's more corrupted than otherwise.

Prior beliefs don't eliminate conflicts of interest

One defense against conflicts of interest is that a person advocated X before he had a conflict of interest, therefore it is now irrelevant that he has a conflict of interest regarding X.

Although this situation might make it more likely that the person's advocacy of X is not corrupted, there is still a conflict of interest to be concerned about. Suppose there is a tradeoff with X and Y at each extreme, and prior to the conflict of interest the person would have chosen a point 70% of the way toward X. Suppose after the conflict of interest, the person chooses a point 90% of the way toward X. Even though the person was pro-X before, the conflict of interest can still make them push harder for X than they otherwise would have.

As another example, suppose the person previously advocated X, and if not for the conflict of interest would have changed their mind to be pro-Y after encountering some evidence. Suppose the person encounters this evidence after they are in a conflict of interest, and therefore continue advocating X even after the new evidence. In this case the person was still corrupted by the conflict of interest, and having been pro-X before doesn't change that.

In the presence of conflicts of interest, how can we arrive at the truth?

The wikipedia page lists three main methods of handling conflicts of interest:

  • Recusal: anyone with a conflict of interest removes themselves from the decision making process. Because so many people have conflicts of interest in the Bitcoin space, this is impractical.
  • Disclosure: those with a conflict of interest make this info freely available. Wikipedia notes that although this is a common mitigation, there is not a lot of evidence that it helps.
  • Third-party evaluations: in the context of Bitcoin, this basically means putting more weight on analysis done by groups or individuals that are not in a conflict of interest situation (or at least in less of one than other participants).

In general, how humans should arrive at truth in an unbiased way is a difficult and unsolved problem. See Thoughts on how to debate productively.