One-member, one-vote is a long-standing practice in co-operatives, ensuring that each member-owner-patron has an equal say in basic governance and that power is not concentrated among those that have contributed the most capital.
But while one-member, one-vote widely distributes influence, it fails to address the intensity members feel for the many activities co-operatives could undertake.
The idea is simple. Co-operatives can pursue many initiatives. Some of these will be highly valued by most people and should obviously be undertaken. But what about the ones that are highly desired or highly disliked by only a few people? Should they be undertaken? Or should the organization instead undertake initiatives that are moderately, but unenthusiastically, supported by a somewhat larger number of people? In short, how do co-operatives (and more generally all organizations) evaluate the trade-offs that underlie the activities they undertake?
Determining these trade-offs is important, as is evident from the long-standing debates at AGMs and around board tables over such things as: the range of products a co-op should offer; guidelines for where credit unions should lend/invest their money (e.g., ethical funds, local versus international); whether co-ops and credit unions should close branches; and whether housing co-ops should invest in new facilities or upgrade existing ones.
The problem with the one-person, one-vote rule is that it is binary — a person is either for an activity or against it; hence it does not allow people to express the intensity of their preferences. While tallying up the votes in favour of activity A and activity B shows, for instance, how many people prefer A to B, it does not provide any indication as to the degree to which A is preferred to B.
Over the last few years, researchers have developed an easy-to-use approach for dealing with the intensity of preferences. This approach is called quadratic voting (see this story in Wired).
The idea behind quadratic voting (QV) is that the cost of voting for something is linked to the square of the number of votes cast. Thus, if one vote for an initiative costs $1.00, two votes would cost $4.00 and three votes would cost $9.00. The point of this mechanism is to make it costly for people to allocate all their votes to one or two things. In contrast to most voting systems, QV provides an incentive for people to distribute their votes around.
QV can be applied to co-operative governance by giving each member a fixed number (e.g., 100) voting tokens. The equal allocation of tokens preserves the idea that each member has an equal say in governance (Posner and Weyl (2014) describe how QV can be applied to general corporate governance where shareholders have differential say).
To capture their preferences, members would use the tokens to cast votes for different initiatives. But in using the tokens, the squared rule is used. As the following hypothetical example shows, a member might cast five votes for one initiative, four votes for two different initiatives, three votes for another two initiatives, two votes for four additional initiatives and one vote for a further nine initiatives. Overall, the member would vote for 18 initiatives and use all 100 tokens.
Allocation of 100 Voting Tokens by a Hypothetical Member Using Quadratic Voting
|Votes Cast||Tokens/Initiative||Number of initiatives||Total Tokens Used|
The example shows that casting more votes for one initiative is increasingly costly in terms of the other initiatives that can be supported. For example, casting six votes instead of five for an initiative requires 11 additional tokens (11 = 36 – 25), tokens that cannot be used to support other initiatives. Thus, unless a member is deeply concerned about only one or two issues, it can be expected that votes will be distributed more equally than would be the case if QV was not used.
As this Bloomberg article outlines, QV has actually been implemented in the Democratic caucus of the Colorado House of Representatives. And it could be implemented in a co-operative or credit union. Since QV works better with more voters and it requires voters to be actively engaged with a variety of initiatives, it might work best with a delegate body. Delegates could be presented with a dozen or so initiatives and QV could be used to determine which ones receive the most votes and are thus most preferred. While the results would not be binding, they would give the board and management a good sense of the delegates’ preferences. And it would give delegates a good chance to see the support there is (or isn’t) for different initiatives.
While QV has some obvious advantages over other voting methods, it does have its drawbacks. One of the best critiques of the idea comes from Tyler Cowen who argues that decision systems should encourage persuasion over trading. I think he is correct — what we need are people working through why some propositions are better than others.
However, I also believe that not all propositions and initiatives can be decided by persuasion since the source of disagreement is often differences in fundamental values or world views. What is needed is some way to determine the intensity of this disagreement, particularly in a world where every issue is becoming polarized and we are more and more retreating to our “echo chambers” and “tribes.” Unlike the one-member, one-vote method, QV has the potential to show the intensity of these disagreements; I suspect that in many cases the disagreements are not as large as has been portrayed. Thus, while QV should not be a substitute for good decision making by those elected and hired to manage co-operatives, it could serve as a useful way of getting more nuanced feedback from members and delegates.
I look forward to seeing if any co-operatives or credit unions pilot the use of QV. If anyone is interested in doing so and would like more information, they should contact us at the Canadian Centre for the Study of Co-operatives.
Posner, Eric and E. Glen Weyl. 2014. Quadratic Voting as Efficient Corporate Governance. University of Chicago Law Review81: 251-272. Click here for a copy.
Image Credit: Wikipedia