# Is it possible to absorb the cost of advertising into the production cost for information goods in situations where attention is scarce?

This is a question about the economics of information goods on the Web.

Let's say that there is an information good that is sold on the Web, that can be produced at some cost of production D, using capital and labor. Notice that this cost of production is only for the first unit, and it costs virtually $0 to make additional copies of the good, since it is an information good. Assume that there is a scarcity of attention, so that without advertising, the good will not be found by many people. Assume that, effectively, the number of the goods sold is proportionate to the advertising budget, but with diminishing returns from advertising. In such a case, is it right to "absorb" the advertising cost into the cost of production? The way I can justify doing this is to say that if I absorb the advertising cost into the cost of production, it is as if I am making additional copies of this information good at some non-zero cost and selling these. In other words, absorbing the cost of advertising into the production cost of the information good makes the information good look like a traditional good (in terms of production costs, especially at the margin). Since there are diminishing returns from advertising, the shape of the cost curves looks similar to the situation for a traditional good where I assume the factors of production to only consist of capital and labor. However, the marginal product of advertising, which is the additional goods sold with every additional dollar of advertising spending, cannot really be connected to the marginal product of capital and labor, so I'm not sure if my approach is correct. • 'is it right to "absorb" the advertising cost into the cost of production' (emphasis by me) Right in what sense? Is this an accounting question? Jul 10 at 8:09 • @Giskard: I mean for the purpose of analysing the behaviour of the firm. Jul 10 at 8:17 ## 2 Answers The cost of advertising is already considered in production, such as when calculating MC=MB. A good example of a familiar product is beer, which has 2 major costs: water and advertising. I'll note that the marginal product of advertising doesn't always have to be in the region with diminishing returns (perhaps when breaking into a market?) but it's not a bad assumption particularly for the applied range. • What do you mean by MC=MB? Do you mean MC=MR? Jul 13 at 14:40 • Yes, Marginal Benefit (MB) can be used interchangeably with Marginal Revenue (MR) most of the time. Jul 13 at 14:47 • Does it matter at all that advertsing sits at a level further removed from capital and labor? As in, it does not affect the balance of the mix of factors of production. That was what was confusing me a bit. Jul 15 at 3:25 • No, that does not matter. Rational actors should observe all costs including implicit ones. Jul 15 at 3:48 • @Joebevo: if it helps you could consider two goods: a good that is advertised and one that is not. To produce the "advertised" good, you actually need to include cost for for advertising. – BrsG Jul 16 at 17:58 Advertising cots are always part of the cost function. When we talk about firm costs $$C(q)$$ in economic model the $$C(q)$$ includes all relevant costs, so advertising expenses are already part of cost function. For example, if firm has to spend \$10 to produce a widget and \$1000 fixed costs to advertise the widgets it sells cost function would be: $$C(q) = 10q + 1000$$ If you also need to spend some extra money on advertising per product, let's say it costs \$2 of advertising spending per widget then the cost function would be:

$$C(q) = 12q + 1000$$

So advertising spending is trivially part of production function. You can adjust this however you like, if you say that your 'information good' itself has only fixed costs (lets say \$100 but in order to sell it you need to spend \$5 per product advertising it then the cost function would be:

$$C(q) = 5q + 100$$.

However, you might want to include advertising in some more nuanced way if this is for some more serious analysis rather than just some trivial exercise, like using Dorfman-Steiner model where advertising cost increases demand and firm optimize their decisions with respect to price and advertising (e.g. see example of that on pp 139 Belleflamme and Peitz) or potentially even some other model.