New Jersey's schools got cheaper, faster internet without a bigger subsidy — by bundling small contracts no provider wanted to risk building infrastructure for

the problem

Individual public schools shopping for internet service one at a time got bad prices and slow speeds, and providers were reluctant to bid competitively on any single small school's contract

background

what everyone would do

The standard response to public schools getting bad internet prices and slow speeds is to increase the subsidy — more federal or state money to offset the high costs schools are already paying. That response treats the price itself as the problem to buy down, leaving the underlying reason providers weren't competing aggressively completely untouched, which means the same high, uncompetitive prices simply get partially subsidized rather than actually reduced.

what they saw

The researchers' account of the program shows New Jersey recognized that the actual constraint suppressing competition wasn't price sensitivity or provider capability, it was exposure risk — a provider's fixed cost of building infrastructure to serve a single small school was hard to justify against the real possibility of not winning that one uncertain contract, so even capable, well-priced providers had no reason to bid aggressively for isolated small contracts. The fix wasn't more subsidy money, it was restructuring the unit being bid on: bundling many small schools into a single large, guaranteed-volume regional contract meant a provider's infrastructure investment was now justified by a much larger, more certain revenue base, removing the exposure risk that had been holding competitive pricing back.

the move

Rather than increase subsidy dollars, the state changed who bid on what — grouping many small, individually unattractive school contracts into four large regional bundles with guaranteed volume, so a provider investing in infrastructure to serve the region could count on winning enough business to make that investment worthwhile, removing the specific exposure risk that had been suppressing competitive bidding.

why it works

By letting providers bid on an entire region's guaranteed demand rather than a single uncertain school contract, the bundling program made the fixed cost of building out infrastructure rational to incur — a provider now knew that winning the regional bid meant serving enough total demand to justify the investment, removing the specific risk calculation that had been suppressing aggressive pricing. This is precisely why the price drop showed up without a meaningfully larger pool of bidders: the same providers who had been bidding cautiously before now had genuine confidence to bid lower, since the underlying business case for investing had fundamentally changed. The finding that most of the benefit accrued to Category D service (the higher-fixed-cost, business-grade tier requiring real infrastructure investment) rather than Category A service (minimal-investment residential-grade) is itself strong confirmation of the mechanism — bundling only helps where exposure risk was actually the binding constraint, and Category A service, needing little fixed investment to begin with, had little exposure risk for bundling to remove. Because the fix required no new public spending, the savings (comparable in scale to the federal E-Rate subsidy itself) came entirely from restructuring the procurement design rather than from any additional public money.

the payoff

Comparing contracts from 2014 (the year before the program) with 2015 (its first year), researchers found the bundling program reduced monthly internet costs by 39% (a $10.32 per Mbps reduction on a $26.78 base) while average internet speeds nearly tripled, from 268 Mbps to 978 Mbps. The researchers found the number of bidders competing in the auctions did not meaningfully increase — the price drop came from existing bidders competing more aggressively once exposure risk was reduced, not from a larger pool of competitors. Most of the benefit accrued specifically to schools receiving 'Category D' service (business-grade internet requiring regional data hub infrastructure) rather than 'Category A' service (residential-grade, requiring minimal investment) — consistent with the exposure-risk mechanism, since only the higher-fixed-cost service category had meaningful exposure risk for bundling to remove. Savings were comparable in magnitude to the federal E-Rate subsidies themselves.

where it breaks

This mechanism depends on the underlying competitive failure genuinely being exposure risk rather than a lack of qualified providers, insufficient subsidy, or genuine cost differences between regions — a purchasing problem caused by something other than fixed-cost risk (a true shortage of capable suppliers, for instance) wouldn't be solved by bundling demand into larger contracts. It also depends on there being enough genuine competing providers willing and able to serve a bundled region in the first place — bundling concentrates purchasing power into fewer, larger contracts, and the researchers themselves flag a real long-run risk: if regional consortiums consistently award contracts to the same single provider over successive bidding cycles, other potential providers may simply exit the market, eroding exactly the competitive dynamic that made the initial bundling effective and potentially recreating a single-provider, uncompetitive outcome by a different route. And bundling requires enough purchasing units willing to coordinate into the same procurement process — voluntary participation worked for New Jersey's willing schools, but units with conflicting requirements, timelines, or governance structures may not be able to bundle demand as cleanly.

what came after

Published as Aryal, Murry, Pal & Palit, 'The Benefits from Bundling Demand in K-12 Broadband Procurement' (NBER Working Paper 33498, 2025; conditionally accepted at American Economic Journal: Microeconomics), the study is cited in public-economics and procurement-design literature as evidence that redesigning how a public institution purchases services can achieve savings on the same order of magnitude as a traditional subsidy increase, without spending any additional public money — while the researchers themselves caution that if regional consortiums consistently award contracts to the same single provider over time, other potential providers may exit the market, eroding the long-run competition the bundling design depends on.

references

1. [1]The Benefits from Bundling Demand in K-12 Broadband ProcurementNational Bureau of Economic Research (Working Paper 33498), 2025nber.org
2. [2]Researchers Credit New Jersey Co-op for Multi-Million Dollar Cost Savings to SchoolsEducational Services Commission of New Jersey (ESCNJ), 2024prnewswire.com

keep it

same kind of clever

• 2009 · Airbnb (Brian Chesky, Joe Gebbia)Airbnb's New York bookings weren't stalled by its algorithm — they were stalled by bad photos, so the founders rented a camera and fixed the listings by hand, one apartment at a time.
• 2012 · Buc-ee'sBuc-ee's turned the gas station bathroom — the part every competitor treats as pure cost — into the actual reason to stop there
• 2006 · Cedars-Sinai Medical CenterCedars-Sinai got doctors to wash their hands by photographing the bacteria growing on their own palms