Wednesday, August 19, 2015

Quality Choice and Net Neutrality



The net neutrality question is one that has been of particular interest of recent, and government policy in this area has many complex repercussions. In short, internet service providers (ISPs) can either provide the same quality service to all content providers at a flat rate (the status quo), or provide a range of qualities and corresponding prices. Much of that debate will not be discussed here. 

The purpose of this post is to evaluate the circumstances of net neutrality in one specific context: where the ISP is an assumed monopolist and can choose provide an array of service quality contracts at differing prices. The consumers then self-select and reveal their preferences based on the contract they choose. It will be assumed throughout that the profit maximizing monopolist has a marginal cost of zero for providing any quality service.

It is clear that should all consumers have the same preferences, the monopolist would choose the one quality/price pair that maximizes profits. If consumers had differing valuations but the monopolist had perfect information on this, then the monopolist would act as a perfect price discriminator (assuming no arbitrage). In either case, the monopolist would charge where the consumer’s marginal value for quality equals marginal cost (zero) and earn the total consumer value. 
             
More realistically, the monopolist does not know consumer valuations. Let’s start by assuming that there are two types of consumers: Low-Types and High-Types. Low-Types value may range from 0-100% of the High-Types and we assume that each group represents half the population. The general result is that when the Low-Types value a good at less than or equal to 50% of High-Types, the monopolist maximizes profits by driving the low-quality good down to zero, or taking classic monopoly profits on the High-Types alone.  

The pair of charts below depicts this situation. Left Chart: the monopolist would like to choose the low-end quality that maximizes their profits (choose X-coordinate at a function's peak). Right Chart: for a profit maximizing quality choice, how much of the total consumer value is being captured.
 
In instances where Low-Types value a good greater than 50% of the High-Types, it benefits the monopolist to keep a lower-quality good in the market. As Low-Types values converge on the High-Types, the optimal low-quality also converges on high-quality. Again, this is the case only when there are an equal number of Low- and High-Types.


We may also vary the distribution of each consumer-type to see how our answer changes. The chart below fixes the Low-Type value to half that of the High-Type, but changes the weighting of group size. The black line is where there are equal numbers of Low and High-Types. One can see that should there be a high proportion of Low-Types, the low-quality converges on the Low-Types value and it is optimal for the producer to attempt a monopoly in that market. 

Instead of charging the High-Types where their marginal benefit equals marginal cost, the preponderance of Low-Types makes it more profitable for the monopolist to extract all the welfare from the Low-Types. Should Low-Types have a value 30% or 70% that of the High-Types, the monopolist's optimal low-quality will be 30% and 70%, respectively, of the high-quality. Not surprisingly, relatively many High-Types tend to drive the low-quality good out of the market so the monopolist can extract the High-Types’ entire welfare. 


Within our context, these results help inform an opinion on the net neutrality debate. The types of quality provided depend on the distribution and intensities of consumer valuations. Where there are many High-Types, or Low-Types perceive the good/service to be far less valuable than the High-Types, the monopolist will tend to drive the low-end quality downwards and often out of the market. 

Large internet content providers, like Netflix, YouTube and Google, tend not only to demand high quality but use the ISP service more frequently as well. Their revenue generally depends on internet-user traffic. The faster the service, the more likely the website will retain or increase traffic relative to competitors. Therefore, these “High-Type” content providers would be keen on acquiring higher quality service if only to not lose market-share.   
Those who are “Low-Types,” perhaps because they cannot afford it, may see their traffic diminish due to low quality and find their relative market position worse than it was before. If low quality service is driven out, then the “Low-Type” content providers will simply leave. This outcome would appear most undesirable. More likely is that there will be a mandatory minimum quality; such a policy would alleviate the worst of these outcomes despite its distortions. On the other hand, if it turns out that Low-Types share of internet traffic is high, or that Low-Types value the service quite highly, then social welfare may be increased by abandoning net neutrality. 

I tend to believe that the vast majority of websites see very low traffic and have low willingness to pay for high quality service relative to the internet giants, suggesting net neutrality may benefit diversity and competition.

No comments:

Post a Comment