All-In Pricing

What it is

An all-in price is a single total. A partitioned price shows the parts and leaves them unsummed. A dripped price shows some of the parts and holds the rest until the buyer is nearly finished. Only the first of the three tells a shopper what the purchase costs at the moment they are deciding whether to make it, which is why disclosure regimes keep converging on it and why sellers keep resisting.

In effect

For most of 2015 StubHub was an all-in seller. Between 19 and 31 August that year it ran a ten-day experiment on randomised halves of its own traffic, with one half kept on all-in prices plus an explicit notice that fees were included, and the other half moved to prices with a fifteen per cent fee revealed at checkout. The concealed half earned about a fifth more revenue, roughly a seventh more people bought at all, and each buyer spent about five per cent more per ticket. Read backwards, that is the revenue a seller gives up by quoting an honest total. On 1 September 2015 the platform switched everyone to fees at checkout.

The laboratory evidence points the same way from the other side. Alexander Rasch and colleagues ran duopoly markets in which one treatment was effectively a ban on dripping, and found consumer surplus higher and seller profits lower when the ban was in force. That is the same trade-off rather than an independent confirmation of its size.

What it does not say

It does not say that an all-in price costs a seller a fifth of revenue in general. The StubHub figure attaches to one platform, one product category, a modest percentage fee, infrequent buyers and ten days in 2015. A flat fee would not have produced the seat-upgrade channel that supplied part of the gain, and the authors of the field experiment decline to offer a welfare verdict at all, naming a two-sided externality they cannot resolve.

It also does not say that all-in pricing solves the problem it addresses. The same paper found that buyers who met a late fee were slightly less likely to leave the platform over the following year, which is the opposite of what a betrayal account predicts, and the authors flag that measure as confounded rather than settling it. What the evidence supports is that an honest total changes behaviour and has a price. What it costs over a longer horizon, and whether buyers are better off, is not what these studies measured.

---