American internet service providers have been the subject of criticism for engaging in usage-based pricing. The latest complaint comes from Massachusetts lawmakers who say the tariff policy based on the amount of data a customer uses harms low-income people, even implying that usage-based pricing will widen the digital divide. Nothing could be further from the truth.
As my AEI colleague Daniel Lyons has explained, usage-based pricing simply reflects a rational strategy for spreading a company’s costs fairly across its heterogeneous consumer base. Flat-rate (“all-you-can-eat”) tariffs are “equitable” only in very limited cases in which consumers have homogeneous (i.e., identical) demands. And this is precisely not what is exhibited in demand for broadband.
The reality is that broadband usage demand exhibits an
extremely wide — and importantly, skewed — distribution. The median for monthly
data usage sits significantly below average consumption. Even though average
consumption has increased over time, this demand pattern has persisted. The
vast majority of broadband consumers use less data than the average — with a
small number of super-users skewing the distance between median and mean usage.
If a flat-rate tariff is set based on average usage, then a
huge majority of lower usage consumers are subsidizing a small number of higher-volume
users. If higher-volume users face no financial cost for their higher usage,
then they will consume even more data. As the high-end users push network usage
figures up, network capacity must increase to respond to demand and so
flat-rate tariffs increase, pricing ever more low-income people out of
broadband markets — even though most of them never even used enough data to
require network expansion. From an income-based perspective, it is clear that
flat-rate tariffs actually work directly against closing the digital divide.
A fairer and more competitive policy is to allow multiple
two-part tariffs that subdivide plans and prices by the amount of data consumed
(i.e., data “buckets”). This is precisely what has occurred in (arguably more
competitive) mobile-data markets. Users can self-select into the tariff
“bucket” that best suits their demands. The network operator can offer many
plans, including some with smaller buckets for low-volume users at much lower
prices than the single average usage-based flat-rate tariff. If price is a
barrier to uptake, then making broadband access even more accessible with low
tariff options will surely help close — not widen — the digital divide.
However, the larger buckets will be priced higher as the “tax and subsidy” from
low-volume to high-volume users is eliminated. High-volume users will now pay a
marginal price commensurate with their higher usage.
Usage-based pricing plays out in two ways. First, network
expansion timing is now determined by the value of demand of the vast majority
of lower-volume users rather than that of the smaller number of data gluttons.
If additional capacity is priced equally to all consumers (via purchase of
additional data “buckets”), then high-volume users in particular will face a
price signal by which to guide their data usage. Thus, usage increases will more
closely reflect the additional value derived from higher data volumes. The cost
of adding new capacity will be allocated proportionately to those who use it
most rather than increasing the subsidy burden on those who are not increasing
their usage by as much. If super-users do not value the data sufficiently to
pay for it at these rates, then network upgrades (and subsequent price
increases) will occur later but more efficiently.
Second, usage-based pricing is a signal of competitiveness in fixed broadband markets. Flat-rate tariffs are sustainable only when market power is being exerted (e.g., a collusive agreement among operators to offer only one tariff type or regulation forcing this to be the case). Because of the inherent cross-subsidy from low- to high-volume users, flat-rate tariffs provide a strategic niche for rivals to exploit when usage is heterogeneous.
A rival can take market share from a flat-rate pricing
incumbent by offering lower usage-based pricing to the majority of consumers
using less data than average. The effect will be a bifurcation of the market
with consumers using less data disproportionately switching to the provider
offering usage-based tariffs. The flat-rate provider now loses the subsidies
(“taxes”) enabling high-volume users to pay lower prices. It will have to
increase the flat rate, reflecting the high data use of the customers that find
them most attractive. Multiple usage-based tariffs will become the industry
Arguably, the adoption of usage-based tariffs is a
reflection of a market becoming more competitive. The greater range of tariffs
emerging is unequivocally good for consumer welfare at the same time as lower
entry prices help to close digital divides. Policymakers should recall that in
both mobile telephony and electricity markets, usage by low-income individuals
was promoted using prepay plans — the ultimate example of usage-based pricing
with a fixed charge of zero and all costs recovered via usage charges. That is,
the diametric opposite of fixed broadband’s flat “all-you-can eat” fees.