Forecasts rather routinely overestimate than underestimate the size of new markets. And that might have spectrum policy implications.
A new study by Aalok Mehta, a doctoral candidate at the University of Southern California, and J. Armand Musey, Goldin Associates managing director, provides a prime example of the forecasting issue, which can lead to inefficiency in the use of scarce spectrum.
Projections of mobile Internet demand produced by Cisco, Coda Research, Yankee Group, and the International Telecommunications Union, for example, have overestimated future demand, they say.
The evidence reveals a persistent tendency to overestimate in both number and value, the authors say.
Of the past seven Cisco mobile traffic forecasts for North America, for example, overestimates were nearly twice as frequent as underestimates (19 overestimates, 10 underestimates), their study suggests.
“Our findings suggest the mobile industry contains much higher levels of inherent demand uncertainty than is commonly estimated and that business and governments may not be fully factoring it into their policy decisions,” the pair say. “To reduce dependence on uncertain estimates, government officials should consider assigning spectrum allocations with greater flexibility of use.”
There are multiple reasons for the forecasting accuracy challenges. Among them: the inability to fully model changes in human behavior based on price changes.
Most goods and services have elastic demand curves. That is to say, as price increases, demand decreases.
So the shift from unlimited usage to buckets of usage, usage caps and overage charges caused behavior to change.
Between 2007 and 2011 period, most mobile Internet access plans were of the unlimited variety. Since then, most operators have moved to capped plans.
So what happened to usage when prices essentially were raised, especially for heavy users? Behavior changed.
Consumption by the top one percent of users dropped from about half of all traffic to just about 30 percent, for example.
The point, the authors say, is that regulators need to remain flexible about spectrum policy. Since demand forecasts can be quite wrong, regulators should periodically reassess whether spectrum originally allocated for one purpose can be better used for other purposes.
Also, since unlicensed spectrum inherently is more flexible than licensed spectrum, a preference for unlicensed spectrum should be the norm, the authors suggest.
Edited by
Maurice Nagle