In the past couple of weeks I’ve been thinking a lot about the problem of status quo bias in the context of achieving meaningful policy change in the electric power industry. If it’s natural for us to prefer the status quo even in the face of beneficial change, how can we overcome those biases that we all have, utilities, regulators, consumers?
Broadly speaking, status quo bias is the tendency to prefer an existing state of affairs to alternative ones. One theory for this preference is that it has its foundation in loss aversion, and that if there’s an asymmetry between an individual’s loss aversion and risk aversion, you can get status quo bias. Individuals having status quo bias in a collective action situation compounds the problem because of their interaction and interdependence of their actions.
The canonical example comes from early experimental work in behavioral economics testing the symmetry of these preferences. Take a situation in which you have coffee mugs and you have a group of individuals (and assume you’ve given everyone the same cash endowment, to reduce the potential wealth effect on the outcome). First you show each person the mugs and ask each person to tell you how much he or she would be willing to pay to purchase the mug from you. Then after they have bought the mugs, you ask them how much they would be willing to receive from you to relinquish the mug. The idea is that if their preferences are symmetric between buying and having, the willingness to pay and willingness to receive should be roughly the same.
But it’s not. Over and over, experimental results reveal that people are much less willing to receive the same payment to give up something as they would be willing to pay to acquire something. In this sense status quo bias can be a manifestation of loss aversion being larger than risk aversion.
How does this concept of status quo bias relate to policy changes? To think more systematically about this question I went back to a classic paper on the subject: “Resistance to Reform”: Status Quo Bias in the Presence of Individual-Specific Uncertainty,” American Economic Review 81 (Dec 1991), pp. 1146-1155. Here Raquel Fernandez and Dani Rodrik presented a model showing that (ex ante) uncertainty about who would be the recipients of the gains and losses from a policy reform could lead to outcomes in which the median voter chooses the status quo outcome, even if total surplus is higher in expectation under the policy change.
What we will show, specifically, is that there is a bias toward the status quo (and hence against efficiency-enhancing reforms) whenever (some of) the individual gainers and losers from reform cannot be identified beforehand. … Significantly, the result holds even if individuals are risk-neutral, forward-looking, and rational and in the absence of aggregate uncertainty regarding the consequences of reform (emphasis original)
Duh, you may say, anyone who has seen the sausage-grinder of policymaking knows that this is true. Yet this is still a powerful result, because Fernandez and Rodrik show that just one simple feature, uncertainty over who gains and who loses, can be sufficient to create enough inertia to destroy the potential achievement of dynamic efficiency gains from policy reforms. Furthermore, they are writing in the context primarily of the trade theory literature, which at the time did not have a good theoretical explanation for why some groups opposed liberalization even when they stood to gain a great deal from it in expectation.
Their model also provides an understanding of how ex ante opposition and ex post support for the reforms can coexist in individuals without having to suppose that their preferences are “irrational”, which helps to explain the success of trade liberalization policies in some autocratic regimes.
Fernandez and Rodrik also make what I think is a very important methodological point about the information environment in which status quo bias arises: when reforms get passed, you get a chance to observe the ex post realization of the outcomes of the reforms, good and bad. But when you don’t pass reforms, you have nothing firm against which to compare the status quo, so it’s easier for you to support staying with the status quo.
There is an important asymmetry between the two cases, however. In the second case (in which a reform is passed and turns out to be unpopular), information is revealed as to how individuals actually fare under the reform. Therefore, if there is ever a second vote or a change to reconsider, the reform may be repealed. In the other case (in which reform is not passed), no new information is revealed, since the status quo is maintained. This asymmetry between the two cases leads to a status quo bias.
This insight creates an opportunity for experimental economics to break through the status quo bias barriers that are apparently inherent features of human nature. Experiments create a realization of the ex post outcomes of reform, in a controlled environment and not out in the real world in real time, that can be compared explicitly to the status quo.
Suppose you could generate some data on the likely outcomes of a reform, in an experimental environment that, although simplified, has the salient features of the policy reform in question. If uncertainty is at the core of the status quo bias, then generating those data (preferably by having policymakers, change agents, and those whose preferences could be decisive at the margin as participants in the experiments) should resolve some uncertainty over the outcomes of the reform. By resolving some uncertainty, if you could overcome enough status quo bias at the margin to swing the balance in favor of reform, then you create the opportunity to unleash the value creation potential of the policy reforms.
Note also that performing an experimental policy analysis as part of the deliberation process may, in a sense, enable some kind of Coasean bargaining or contracting over anticipated gains and losses. Such a process could enable us to implement something that we economists are used to thinking about as only a blackboard tool — the Kaldor-Hicks lump-sum transfer. Basically the argument is that if a possibility for a lump-sum transfer from winners to losers exists, as long as the net effect of the change is positive then the move is Pareto efficient (and should take place in a frictionless and transaction-cost-free world, such as the blackboard).
In other words, we should think about experimental policy analysis as a tool for reducing transaction costs that can hamper the movement from a status quo policy regime to one that promises great aggregate benefits, but where we have uncertainty over who gains and who loses.
That argument assumes that Fernandez and Rodrik’s theory is correct, and that ex ante uncertainty is the root cause of status quo bias in the policy arena. I think that’s a pretty fair assumption.
These ideas point to a way forward. Recognize and acknowledge status quo bias, without judgement. As a preliminary step in deliberating policy change, conduct experimental analyses with potentially decisive stakeholders, evaluating their positions before the experiments and after. Construct experiments that the decisive stakeholders perceive as sufficiently capturing the salient features of the policy change, so that they trust the process and the results of the analysis are credible, and therefore more likely to be persuasive.
There must be more, though. This is just a beginning.