Pricing and Marginal Cost

Regarding "predatory pricing," Netscape 7.0 is alive and well, complete with tabbed browsing.
Microsoft has fierce competititon in the browser market, such as Firefox and Opera. In fact, the Soft One has been losing browser market share for quite a while.
I can't think of a single case in which "predatory pricing" was ever bad for consumers.

Good question and great example. People who see competition as necessary for their to be a market would say at least 2 -- so that there is a choice. If you're familiar with the San Francisco Bay Area, it's easy to visualize the silliness of that position. In that case, sure, the Bay Bridge has competition -- if you want to drive clear down to the San Mateo bridge in Oakland and San Francisco traffic. Of course, there is competition from ferries and BART as well, much as cable provides competition to telephone. Again, great example.

I think the social optimum has more to do with capacity than number of bridges. If the total capacity is such that congestion can be mitigated by pricing signals, then we have the social optimum. If high congestion pricing doesn't reduce traffic to reasonable levels and high congestion is a recurring and frequent problem, then the system could use more capacity. People might even be willing to pay higher non-congestion pricing so that congestion happens less often.

Doesn't the optimal investment in bridges have to be where the "reduced congestion" benefit from the next bridge no longer exceeds the cost of the bridge? Seems like a basic question of equating marginal cost/benefit or am I missing something?

Marginal price does seem to tend to marginal cost, even when it's close to zero - aren't digital products (absent itunes, and we'll see about that) pretty much all turning to fixed subscription pricing?

Does Anyone see the fallacy of the above argument. The Costs of Bridge construction and maintenance will be the minimum Cost of the Toll, while the variance of Toll rates between Peak and Slack will reflect the Expense of conjestion. lgl

Isn't this a bit of a straw man though? Those advocating marginal cost = 0 would have a two part rate. You pay a lump sum then you cross for 'free' during off peak.

I guess if you look hard enough you can find a non-zero marginal cost for driving over the bridge during off-peak hours.

The capital cost for borrowing money to build the bridge is measured per unit of time (some amount of interest per year).

So if it takes some amount of minutes to drive over the bridge, assuming during non peak times you are the only car on the bridge, the capital cost for that amount of time is a start for the marginal cost.

If one person drives the only car that goes over the bridge in one day, then in some sense the marginal cost for that car is the cost of keeping the bridge open for one day, including the capital cost that keeps the enterprise from going bankrupt.

The marginal cost goes down neatly from one car a day until the point where there is more than one car on the bridge at the same time. If there are not drivers are willing to pay that, there are too many bridges.

We're leaving the region I'm comfortable trying to model when we start talking about multiple cars on the bridge. (If we haven't already left, or if were were ever there.)

At this point drivers can auction the right to be on a bridge with fewer cars.

If the bridge operators were to set a limit or cars per unit of time in advance, they could collect a scarcity rent - above the marginal cost.

What limit maximizes bridge operator profits? That depends on the demand characteristics of the drivers.

The social optimum is pretty easy. No scarcity rent. Given good enough capital markets, enough bridges are built that drivers just pay their per-time shares of the capital costs.

Without good enough capital markets, the amount of available scarcity rent (i.e. congestion) that signals that it will be profitable to make a new bridge depends on how not good enough the capital markets are.