IO Chapter 6

A review of competition: As in the case of monopoly power, the extreme case - perfect competition - is rare.

What are the assumptions of perfect competition?

1. many suppliers (atomicity)

2. homogeneous product

3. perfect information

4. equal access to technology

5. free entry (no barriers to entry)

Although the book lists these as 5 attributes, they are connected. In addition, firms work very hard to overcome 2 and 4 (which are also related), since they want to have name brand recognition/a unique product and as such a chance for some monopoly power.

What were the mechanical realities of PC?

pq - C(q) = p, since p was given (firms are price takers) MR=MC=p

allocative and productive efficiency are achieved. There is no dead weight loss in the short run and in the long run the most efficient firms are the ones that stay in business. These are issues of static (unchanging) efficiency, but the reality is that production is dynamic. In particular, technological and innovation is dynamic. As such, we need to also think about dynamic efficiency.

How does PC hold up under empirical scrutiny? Not very well in most markets.

What predictions come out of the PC model?

1. There will be entry or exit until those firms with the lowest cost function (the most efficient) remain. In the long run (LR) we reach the min. of the LRAC curve. We can show this graphically.

2. In addition, in the LR profits will be driven to 0 by exit (when profits are <0) and entry (when profits are >0).

3. Plant size will be homogeneous (if costs, technologies and products are homogenous) and yet this is not the reality.

4. On firm size the predictions are vague. On the one hand, having multiple plants could be suboptimal if the managerial costs of overseeing multiple plants are positive. In this case there should be no incentive to have multiple plants. On the other hand, if the costs are zero, no optimum size is predicted. (The reality may be neither, instead there may be advantages to multiple plants.)

These are all four questions which economists have examined empirically. The evidence contradicts all 4 conclusions.

1. Exit and entry occur simultaneously.

Although one might think it is not surprising that some entry and exit might be happening simultaneously, it is the size of the simultaneous entry and exit that is surprising. For example, suppose you have 1000 firms, and 200 drop out at the same time that 150 new ones join in. Although the net position is only -50 ( 5%), 20% of firms are exiting while another 15% are entering. One might expect one pattern to swamp the other (200 (20%) drop outs and 10 (1%) new , for a net of 810 (19% drop) but this doesn't appear to be the case generally.

Looking at Table 6.1 on p. 88 we can see what has been happening in terms of entry and exit. In which case is the difference btw exit and entry the largest/smallest? Can we make any predictions about whether an industry is making super profits or losses from these data?

Korea and Portugal?

Definitional aside: what is a 4 digit industry? We will go over this in class at some point.

2. Firms make large profits even in the long run.

Research by Dennis Mueller suggests that firms do make positive profits, and often large ones, over long periods of time. (Can we get some #s?)

3. Size of plants varies. ??

4. Size of firms varies, but there is evidence of increasing, rather than constant returns to scale. ?? For instance cross country comparisons suggest similarities across the US, Japan, France and the UK. (p. 89, figure 6.1)

Large firm size is associated with less a chance of exiting, but also less of a chance of growing. As such, small firms either grow quickly or exit quickly. This suggests that there are certain returns to scale.

Economists have come up with various models which maintain some but not all of the PC assumptions. For instance the model of competitive selection assumes that firms are price takers, have a homogeneous product and have perfect information about prices, but that they do not have access to the same technology and that they face barriers to entry. This model allows entry and exit at the same time, and allows some firms to make large profits but remains efficient, (I do not find this model very realistic or useful, so we will not spend a lot of time discussing it. Notice that the author does not provide an empirical example. Although one can come up with various theoretical models which relax one or more of the PC model assumptions, and provide an alternative explanation of how markets operate, it seems to me that the test as in the case of the PC model of whether a model is worthwhile or useful is whether the empirical evidence supports these models.)

A more useful model is monopolistic competition. In this model, firms produce heterogeneous products. Ex: Shampoo. Can you think of others? Clothing (perhaps, although at the retail level it is getting quite concentrated.) It is useful to compare the monopolistic competition model with both the PC and dominant firm models to see how it is similar/different.?

Which of the PC assumptions hold/are violated? Firms are no longer price takers. Because of their unique product, each firm faces a downward sloping demand curve and can set their own price. But at the same time, because there are a number of close substitutes, the sales of each firm will depend not only on how they price their product but what their competitors do. A large number of firms, perfect information, technological access and exit and entry can all still hold. What about profits? In the short run firms make a profit because they set MR= MC and then read the P as a monopolist would, with P>AC. But in this case, because of free entry, firms will keep entering the market, thus reducing each firm's demand curve. Eventually market will end up where P=AC, but this will not be at the cost minimizing point. This result is not allocatively efficient, since p firms will end up where p=AC, but not where p=MC, and as such they do not minimize their cost function.

See graphs on p. 92/3 Figures 6.2 and 3.

p > AC implies that firms are making profits, but they could be efficient.

p > MC implies that firms are not operating