Equilibrium is a central concept in economics. But what is multiple equilibria? In high school or even university economics courses, students may not ever encounter multiple equilibria though the concept, in my view, is very useful. Let me introduce some very rough idea about it in this post.
First, students encounter equilibrium this concept almost only when they learn demand and supply analysis. In such an analysis, equilibrium is the intersection point of the demand curve and supply curve. Other than this, perhaps they may not know more concretely what equilibrium is about.
In fact, equilibrium may be better understood via adjustments or movements. In equilibrium, no (further) adjustment will take place, or no movement away from it is expected to take place, given that no new factors appear. Not in equilibrium, adjustment will, or is expected to, take place. The reason why the demand-supply intersection point is an equilibrium is exactly this. At the intersection point, demand equals supply and so price will not adjust. Not at the intersection point, either quantity demanded exceeds supplied (so price will go up), or quantity supplied exceeds demanded (so price will move down). In other words, adjustments will take place.
Of course, this example is only about market equilibrium. There are equilibrium concepts other than market equilibrium in economics. For example, game theory uses the concept called Nash equilibrium, which is often not about the whole market but an intersection result in small group of people. Anyway, equilibrium is normally related to the choice of more than one person. These people each choose to do something, given what others will do or have done. Equilibrium is a state where all people would rather stop but not move away from it.
Normally, the demand curve and supply curve will intersect at only one point, not more than one point, as demand curve is downward sloping while supply is upward sloping. So, this is a single or unique equilibrium case, not multiple equilibria. But you can easily imagine that there may be some special cases in which demand and supply intersect at more than one point.
Consider this example. Recall a past post: labour supply curve is often found to be backward bending. This means the supply curve is upward sloping when price (wage) is low, but is downward sloping when price (wage) is high. Meanwhile, the labour demand curve is downward sloping. Then, it is possible that labour demand and supply curves intersect at two points: one at where supply is downward sloping, and one at where supply is upward sloping. (You may draw a diagram to show this.)
The fact that there could be two equilibrium points in labour market does not mean that both equilibria will be realized. In fact, what is realized is only one point. But both points could possibly be realized. Which point will be realized? This is a difficult aspect.
Economists normally use the equilibrium concept to explain what things will happen. They say things will happen in equilibrium (so price will be determined in the demand-supply intersection point). But now if there are more than one equilibrium, economists need something extra to tell what will happen, or which equilibrium will be realized. One obvious hint is the so-called initial condition: which equilibrium will be realized depends on what is the staring point. For example, if price starts at a low level, the intersection point of demand and the upward sloping segment of supply will be realized. If price starts at a high level, the intersection point of demand and the downward sloping segment of supply will be realized.
Now there is a very important implication that economists can draw from such a multiple equilibria analysis. First, if there are two (or more than one) equilibrium points, perhaps we can tell which one is better. For example, a high-wage equilibrium may be considered better (for workers) and a low-wage equilibrium may be considered worse. Second, if both good and bad equilibrium points may be realized, we may not need to accept what has been realized (bad equilibrium) but can try to do something to realize the good equilibrium. Third, since equilibrium is self-enforcing (inducing adjustments when not in equilibrium) once the conditions are right, moving the equilibrium from bad to good need not involve great efforts (or high cost). All we need to do may simply be to change the initial conditions (or some other conditions). Once the adjustment process takes place towards the good equilibrium, no further effort is needed.
So, if multiple equilibria exist in the real world, policies to improve the world may not be as difficult or costly as we originally think. The question is whether multiple equilibria really exist.
Yes, we will be happy if the world can be improved at low cost (when multiple equilibria exist). But, as dismal scientists, economists are normally skeptical of the existence of a wonderful world. Some economists indeed have doubts on the multiple equilibria concept. I don't want to go through their arguments in this blog post. I just want to let you know that this concept is still a little bit controversial in the economics circle. If we want more people, economists in particular, to embrace it, we need some or more convincing examples. The labour market example above is not such an example. This is only an example, I think, students can easily understand and so useful for illustrating the multiple equilibria concept.
I encountered multiple equilibria concept many years ago when I learned some development economics. That's the first time I learned this concept. The idea is this. We know in the real world many countries are poor but many are rich. Not all poor countries are lack of resources but they are still poor, just that they fail to develop. Some rich countries are also lack of resources but they are still rich, just that they can develop. Why? Of course, there could be different explanations. But one is multiple equilibria. The poor countries are trapped in a bad equilibrium while the rich countries can reach the good equilibrium. I was amazed by this idea the first time encountering it. But then I know many economists are skeptical, and for development, there could be different convincing explanations. They need not choose the multiple equilibria hypothesis if they can choose others.
Are there convincing examples in which multiple equilibria is almost unavoidable? I think there is. The example is convincing because we, with or without learning multiple equilibria, have already thought in the same way. The example is bank runs.
What is the crucial issues involved in bank run?
First, the reason why there may be runs on banks is that banks receive deposits but deposits can be withdrawn at very short notice while banks lend money to businessmen on a much longer term. If deposits can't be withdrawn at short notice, without the flexibility, people may not want to deposit money in banks (why don't they invest in something else). If banks recall loans at short notice, the businessmen might bankrupt, the investment projects undertaken by the businessmen may not be mature enough to generate returns, and so such a recall will not give banks much.
Second, banks will keep only a small percentage of their deposits as reserves. The rest will be lent for earning interest income. So, if too many depositors withdraw at the same time, banks never have sufficient reserves to cope with these withdrawal demands. If banks recall loans that support investment projects that are mature, banks can't get back too much and will make loss. So, the crucial thing for the banking business to work is that depositors won't withdraw at the same time.
Third, if depositors believe that a bank works well, there is no reason to withdraw immediately. But even if a bank work well, when many depositors withdraw, the bank still doesn't have sufficient reserves to return the money. They may be forced to recall a long-term loan back and makes a loss. When the loss (unnecessary if no withdrawal happens) is realized, banks may not be able to return the deposits to someone who withdraws it too late. So, it is wise to withdraw early if others are also in a rush to withdraw.
The result is well known: either all depositors are calm and won't withdraw at the same time, or all depositors fear and withdraw at the same time (bank run).
We, economists in particular, all know that this is what banking business is about. But before 1980s, economists are not aware that this is exactly about multiple equilibria. In the good equilibrium, no one withdraw in a rush; in the bad equilibrium, depositors run on their banks. Douglas Diamond and Philip Dybvig have written a paper on this in 1983 (the Nobel lecture can be found here). It is this paper that makes them win the Nobel prize in economics in 2022.
No comments:
Post a Comment