Telemarketers and Externalities

The discussion of the national Don't Call Registry (see "Don't Call Me Baby" below) has raised some interesting questions for me, particularly about the nature of the externality involved. When a telemarketer calls me, and I don’t buy, we both pay a cost, and both of us would rather have avoided the encounter. When the telemarketer calls someone who buys, they both gain from the call. The problem is that in order to find the person who wants to buy, the telemarketer has to call a lot of people who don’t. Overall, the buyers benefit from telemarketing and I lose; in other words, the buyers are imposing the externality on the non-buyers.

So, how to set up the initial allocation of property rights? We could let telemarketers call anyone they want, call anyone except those on a “don’t-call” list, or not call anyone except those on a “do-call” list. My rough impression is that those that buy from telemarketers are few in number but get a substantial gain. For non-buyers the amount of annoyance (per person) is small but there is large number of non-buyers. This argues for the do-call list. The benefit to the few buyers would be enough to cover the cost of getting on a list, while the costs of annoyance for the non-buyers might not.

If don’t-call lists just remove non-buyers, then why are telemarketers against this? I think part of the reason is that there is not just heterogeneity of consumers, there is also heterogeneity of products. I might buy some products but not buy others. When deciding whether to go on a list I’ll weigh the benefit of fewer calls against the expected gains from the products I would buy. The fact that I go on a don’t-call list says something about my attitude towards telemarketed products on average, not any specific product.

A little formalism: Say that the probability of a sale is p, the benefit to both the buyer and seller from the sale is x (to make things easy), the cost of a call is c, and the annoyance of a call is a. The telemarketer will call as long as px > c, while you will want them to call as long as px > a.

When I choose to be on a don’t-call list I am making a decision based on the expected value of the companies that will call based on my preferences. This could cause telemarketers to be against don’t-call lists for several reasons. As long as c is below a, there are companies that would want to call me (in an expected value sense) even though I would not want to be called by them (for email advertisements, c effectively equals 0, which is why everyone hates spam). Companies with high values of p would also want to call me, and I would want them to call, it’s just that I can’t limit my calls to only them. Don’t-call lists are efficient in separating low-p people from high-p people, but are not good at separating low-p products from high-p products.

When listening to a telemarketer I came up with this plan: Have telemarketers hooked up to small electrodes. If they call, and you don’t buy their product, you can press a number and give them a small shock – nothing dangerous, just enough to smart. Telemarketing companies would have to pay their workers a higher wage to compensate for the shocks. The worse the product, the higher the cost. Borderline companies will no longer find it profitable to telemarket. For good companies, the higher wages are passed along as higher prices to their customers (but they are the beneficiaries of telemarketing in the first place). While the non-buyers are not directly compensated, hopefully the incentive effects would outweigh the other costs.

Of course, I’d also be open to market-based plans that don't rely on electricity.