Income and pickup time set how many parcels pile up in a US apartment building

Most US apartment buildings were planned before the parcels arrived: two in three were built before 2000, when online shopping was under 1% of retail sales, against about 17% today (1, 2, 3). Their parcel space was fixed then. For new buildings, the only US rule, from the US Postal Service (USPS), counts homes: about 20 parcel lockers per 100 homes, whatever the rent (4). Yet affordable and high-end buildings don't receive the same number of parcels. This article argues that income mix belongs in the sizing too.

In 2022, households earning $200,000 or more reported 2.8 online goods deliveries a week (parcels of purchased goods, not food or grocery orders), against 1.0 for those earning $25,000-35,000, and deliveries rose at every income step in between (5). The figures come from the National Household Travel Survey (NHTS), the federal survey of how Americans travel and shop. We use its renter households, the closest match to apartment residents; owners show almost the same pattern (5).

The second input is pickup time, which decides how many parcels sit in the building at once. In the one independently measured US building, parcels waited about half a day on average in 2020, when many residents worked from home. At that speed, in the busiest holiday week, a building of households earning $200,000 or more holds about 37 parcels per 100 homes, nearly twice the 20 lockers USPS requires; at $25,000-35,000 it holds about 13. If parcels wait a full day, which we assume rather than observe, both double (4, 6, 7).

Parcels in the building at the same moment in the busiest holiday week, by income and pickup time

HomesPeriodBusiest holiday weekDecemberNormal weekShare

Our calculation: parcels arriving a day x hours each waits, from 2022 renter goods deliveries by household income. December runs 1.35 times a normal month (Luxer One), the busiest week 1.81 times a typical week (NMHC). Monday spikes and oversized parcels are left out. 2022 rates; add about 8% for a building opening in 2026.

The 12-hour line is close to the one measured mean, 12.5 hours in Seattle (7); the 48-hour line sits just above the slowest measured case, 41 hours at one apartment community (12). Parcels handed over as they arrive need almost no space to wait in. For a building with mixed incomes, take the average of its households' lines.

The USPS line assumes one mailbox compartment per home: the rule asks for at least one parcel locker for every five compartments (4). Our sources don't say how many parcels a locker holds, or whether carriers other than USPS may use it, so the comparison shows scale, not capacity.

Deliveries rise at every income step, from one a week to nearly three

Parcel load per home rises at every step up in income. The figures rest on what survey households report, and an independent analysis of the same survey supports them.

Deliveries rise at every step up in income

Renter households, 2022. All types add food, grocery and service deliveries. Each point runs to the next ($25k = $25,000-34,999); $100k covers $100,000-149,999.

No gate log, operator count or platform figure in our sources splits deliveries by income. An independent regression on the same survey counts adults rather than households and holds other factors fixed. It finds that adults in $100,000+ households receive about 1.9 times the deliveries of those under $25,000 (13). The Federal Highway Administration (FHWA), which runs the survey, and a peer-reviewed study of three survey waves also find an income gap, one that is narrowing (14, 15).

Counting all delivery types (adding food, grocery and service), every income step receives more, but the climb is about the same: 1.4 a week at $25,000-35,000 against 3.9 at $200,000 or more (5). Between 2017 and 2022 the ratio between the broad groups fell from about 3.5 to 2.5, though the absolute gap widened (see More from our data) (5).

Where a building stands matters too. Most US counties have a median household income between $50,000 and $75,000, but the map ranges from rural counties under $40,000 to suburban counties above $150,000 (16). A building's parcel load follows the incomes around it.

Median household income by county, 2024

All households. A county is an average of buildings at every income. Connecticut, now reported by planning region, is grey.

The levels fit what operators see on average, though operators count parcels in their own ways and years. The average renter household reports 1.30 goods deliveries a week (5). Property managers surveyed in 2018 reported about one parcel per home a week; Luxer One, a parcel-locker vendor, reports 2.1 for its 2025 client base (9, 17). A building of only high-income renters sits above that range, and counting every delivery type, 18% of renter households already receive 3.1 or more a week (5).

On the one rent comparison available, deliveries rise faster than rents

Our sources have no rents by household income. The nearest stand-in compares Class A apartments (the newest, best-equipped tier) with Class C (older, plainer stock): effective asking rents for new leases in Jan-Mar 2024 were $2,213 a month against $1,347, about 1.6 times (19). Goods deliveries rise about 2.8 times from the under-$50,000 group to the $100,000-plus group, so parcel load climbs faster than rent; the stand-in compares buildings rather than renters, so this points against a flat allowance without settling it (20).

Deliveries doubled from 2017 to 2022; since then, parcels per household have grown about 2% a year

Deliveries to US households doubled between 2017 and 2022, then settled into a slower climb. 2022 rates therefore understate the load on a building opening now, but by much less than the jump before 2022 might suggest.

US households went from about 1.1 deliveries of all types a week in 2017 to 2.4 in 2022, owners and renters alike (5). The Federal Highway Administration, which runs the survey, counts the same jump per person: from 2.5 to 5.4 deliveries a month, up 117%, at every income level (14). No later survey counts household deliveries. The best count since comes from Pitney Bowes, a shipping company that publishes a national parcel index: 162 parcels per US household in 2022 and 171 in 2025, about 2% a year by our calculation (10, 11). Online spending grew faster, by 5.4% in 2025 and 12.2% year on year in Apr-Jun 2026, though sales are counted in dollars, not parcels (3, 21). Luxer One, a locker company, reports 15% more deliveries across its properties in 2025, but that counts its own growing client base, not the country (22).

At about 2% a year, 2022 rates understate a building opening in 2026 by roughly 8%: about 101 goods deliveries per 100 homes a week at $25,000-35,000 become about 109, and 284 at $200,000 or more become about 307 (our calculation) (5, 10, 11). That assumes equal growth across incomes. Weekly online grocery ordering among lower-income adults rose faster after 2022, so lower-income buildings may be understated (23). If the 2026 rise in online sales lasts, the gap would be larger.

After 2022, parcel counts grew about 2% a year; online sales faster

Indicator	What it counts	Period	Change
Parcels per US household (Pitney Bowes)	All parcels shipped, business and returns included	2022-2025	162 to 171, about 2% a year
E-commerce sales (Census)	Retail sales value	2025 vs 2024	+5.4%
E-commerce sales (Census)	Retail sales value, seasonally adjusted	Apr-Jun 2026 vs Apr-Jun 2025	+12.2%
E-commerce share of retail (Census)	Share of retail sales	2023 to 2025	15.3% to 16.4%
Deliveries at Luxer One properties	One vendor's own buildings, a growing client base	2025 vs 2024	+15%

Each counts something different; none is split by income. About 2% a year: our calculation.

Parcels waited about half a day in the one measured building

In the only independently measured US building, most parcels left within hours. We plan with 12 to 48 hours; that range is our choice, because no study measures pickup across building types.

The building is a 26-story, 133-unit condominium in Seattle's Belltown, with a lobby locker (24). The data run from June to December 2020, during the pandemic, and the authors note that working from home may have shortened pickup times (25). Across 6,823 packages, the median pickup time was 4.2 hours and the mean 12.5 hours (7). Of those, 84% were collected within a day and 99% within a week, with no storage limit or late fee (7). Pickups were slower at weekends and in autumn and winter, and the mean rose 19% from June to December (7).

The other figures are thinner:

• A Luxer One case study of one 337-unit apartment community reports a mean of 41 hours in a single month, rent tier unstated (12).
• In university housing using the same vendor's lockers, over half of packages left within 12 hours (26).
• Australia Post's public lockers show dwell times of 15-16 hours across more than 867,000 pickups (27).

Seattle's mean is almost exactly our 12-hour line; our 48-hour line sits just above the vendor case.

Pickup time drives space through Little's law, a standard queueing result: parcels on hand equal parcels arriving per day times the average days each waits. Double the wait and you double the slots. Pickup also slows in winter, just as volume peaks, so the two compound at the worst moment.

In the busiest week at 12-hour pickup, buildings from $50,000 up reach the USPS minimum

In the busiest holiday week, at 12-hour pickup, a building of households earning $25,000-35,000 needs about 13 storage positions per 100 homes (9-18 within the survey's margin of error), within the USPS minimum of 20. From $50,000 it reaches or passes the minimum: about 22 at $50,000-75,000 (17-27) and 37 at $200,000 or more (28-46) (4, 5, 6). In a normal week, even the highest-income building fits within the minimum at that speed, at about 20. Each storage position holds one parcel; a USPS locker may hold more or fewer, and may serve USPS only.

Storage needed in the busiest holiday week depends on pickup time as well as income

PeriodBusiest holiday weekDecemberNormal week

Our calculation, per 100 homes. 12 hours is close to the one measured mean (Seattle, 2020).

USPS announced in April 2021 that it had tightened the ratio from one locker per ten mailboxes to one per five, to cope with package volume (28). It sets no maximum and does not vary by rent (4).

Moving from 12 to 48 hours changes storage four-fold, but that follows from the range we chose; it is not measured (6). The income gap is measured: about 2.8 times from $25,000-35,000 to $200,000 or more, and about six times from under $10,000 to $200,000 or more (5). If real pickup times vary as widely as our range, pickup matters about as much as income. Income comes with the market a building serves; pickup time may be partly within its control, though no study in our sources shows by how much.

Work it out for your building

HomesIncome$25k-35k$50k-75k$100k-150k$200k+PeriodBusiest holiday weekDecemberNormal weekShare

Goods deliveries a week

Goods deliveries130-250

Storage positions, if parcels wait

12 hours9-18

24 hours19-36

48 hours37-71

Our calculation: 2022 renter goods deliveries by household income; ranges are the survey's 95% margin of error. December runs 1.35 times a normal month (Luxer One), the busiest week 1.81 times a typical week (NMHC). 2022 rates; add about 8% for a building opening in 2026.

The objection: the scenarios could be too high or too low

The objection cuts both ways. Post-2022 growth and short spikes push real needs up; fast measured pickup and mixed tenancies pull them down.

Upward: our volumes are from 2022, and growth since is uncertain. The busiest-week factor is a weekly average, so Monday spikes and oversized items come on top (29). In the Seattle building, daily deliveries nearly doubled, from 23.9 in June to 46.1 in December, though June is a quiet month rather than an average one (25). Nationally, ShipMatrix, a parcel-industry analyst, expected average daily volume in the 2025 holiday season to rise only by low double digits, after a 50% rise in 2020 and nearly 100% in 2013 (30). And if a USPS locker holds one parcel or fewer, or serves only some carriers, a building meeting the minimum has fewer usable positions than our figures count.

Downward: the busiest-week factor comes from what property managers reported in 2018, not from counts; on the December factor every figure falls by about a quarter (8, 9). The only measured mean pickup is about half a day, not one (7). Few buildings sit in one income bracket: a building split evenly between $25,000-35,000 and $200,000+ households needs the average of the two, about 25 positions per 100 homes at 12 hours and 50 at 24 hours in the busiest week (our calculation) (6). The one measured peak fits the 12-hour range: on 12 December 2020, the Seattle locker held about 28 parcels, about 21 per 100 homes (our calculation) (25, 31). The same building also received fewer parcels than our figures give. In December 2020 its locker took 46.1 deliveries a day (25); applied to its 133 homes, our December factor gives about 73 a day for households earning $200,000 or more, and the busiest-week factor about 98 (our recalculation: 2.84 goods deliveries a week per home ÷ 7 × 133 × 1.35 or 1.81) (5, 8, 9). That is a flag, not proof that our figures run high, because the comparison is not like-for-like: the residents' incomes are not reported, the locker counts physical parcels while the survey counts self-reported deliveries (one order can arrive as several boxes, or several orders in one box), the data are from 2020 rather than 2022, and downtown condominiums may hold smaller households than the national average.

We concede the evidence can't say which way real buildings fall. The Seattle speed comes from one condominium in 2020, when working from home may have sped up pickup (25). The two measured means, 12.5 hours there and 41 hours in the vendor case, are about 3.3 times apart (our calculation) (7, 12), close to our planning range, but that doesn't show which buildings sit where. No study in our sources measures pickup in affordable buildings or staffed parcel rooms. The figures are a planning frame; a building's own pickup log is the better number.

What it means for developers and operators

Size parcel storage from the number of homes, the tenants' income mix and the expected pickup time, and treat pickup speed as part of the design.

For property developers:

• Start from the planner: the 12-hour line for buildings with fast notification and pickup, the 24-hour line otherwise, in the busiest week. Add headroom for growth since 2022.
• Treat the USPS minimum as a regulatory floor, not a capacity plan.
• Put handover points near the entrance. USPS already requires its lockers to sit in vestibules, halls or lobbies, clear of swinging doors (4).
• Don't assume affordable buildings need little space: households earning $25,000-35,000 still receive about 100 goods parcels per 100 homes a week (5).
• Plan the space early. In a 2018 survey by the National Multifamily Housing Council (NMHC), the apartment industry's trade body, a quarter of communities without parcel lockers said they lacked the space, and another 22% had no good place to put them (9).

For logistics operators: a lobby locker halved couriers' time inside one Seattle building (32). A locker full at the peak sends drivers back to door drops, losing the saving when it matters most; that last step is our reasoning, not measured.

More from our data

At the extremes, renter households earning $150,000 or more received 3.75 deliveries of all types a week in 2022, against 0.85 for those under $10,000; in 2017 the figures were 2.37 and 0.32 (5).

The income ratio narrowed, but the absolute gap widened: high-income renters received 1.40 more deliveries a week than low-income renters in 2017 and 1.84 more in 2022. For owners the gap went from 1.47 to 2.28 (5).

A tax-credit project is one financed by the Low-Income Housing Tax Credit, the main federal subsidy for affordable rental housing. An 80-home building, the average tax-credit project size for 2000-2019, of households earning $25,000-35,000 receives about 81 goods deliveries a week, about as many as a 28-home building of households earning $200,000 or more, at about 80 (our calculation) (5, 33).

How we did this

Delivery rates come from our weighted analysis of the 2017 and 2022 NHTS public-use files (FHWA and Oak Ridge National Laboratory) for renter households, and for all households in the growth figures; the code is saved with the data. Missing delivery answers are rare: imputing them changes no estimate by more than 0.07 a week (5). The 2022 response rate was 11.3%, down from 15.6% in 2017, and the survey moved almost entirely to web response, so the waves are not strictly like-for-like (18, 34).

Our calculations, listed as "Our calculation" in Sources, are the storage positions (including the evenly mixed building), the income and rent-gradient ratios, the 2026 uplift, and the growth and pickup ratios. Pickup times of 12, 24 and 48 hours are scenarios. The December factor is one vendor's 2025 seasonality, and the busiest-week factor a 2018 survey of property managers, both applied to 2022 volumes.

Terms used:

• Low- and high-income groups: households earning under $50,000 a year, and $100,000 or more; used where a source splits income only this way.
• Goods delivery: an online purchase of goods brought to the home. Food, grocery and service deliveries are left out because they don't use a parcel room (18).
• Storage position: one locker cell or shelf slot holding one parcel; our unit, not the USPS's.
• Pickup time: the hours between a parcel's drop-off and its collection by the resident.

Missing from our sources:

• measured hold times in affordable or staffed-room buildings, or across building types;
• how many parcels a USPS parcel locker holds, and which carriers may use it;
• median rent by household-income band.

Sources

1. NMHC Quick Facts: When Were Apartments Built? (NMHC tabulations of 2024 American Community Survey microdata, US Census Bureau; updated 6/2026), 2024: https://www.nmhc.org/globalassets/research--insight/quick-facts/apartment-stock/when-were-apartments-built.xlsx

2. Resourcera compilation of US Census Bureau annual e-commerce data (original: Census Annual Retail Trade Survey, E-Stats Table 5 Historical 1998-2009), 2000-2020: https://resourcera.com/data/retail/us-ecommerce-sales/

3. US Census Bureau: Quarterly Retail E-Commerce Sales, 2nd Quarter 2026 (CB26-133), Q2 2026: https://www.census.gov/retail/mrts/www/data/pdf/ec_current.pdf

4. USPS: Handbook PO-632, Section 5-1 Apartment Communities, current in USPS handbook PO-632 (change announced April 2021): https://about.usps.com/handbooks/po632/po632_05_001.htm

5. 2022 NextGen NHTS public-use files (CSV): household and person files (FHWA / Oak Ridge National Laboratory), our analysis, 2022: https://nhts.ornl.gov/assets/2022/download/csv.zip

6. Our calculation from sources 5, 7, 8 and 9, 2022 volumes; 2018 and 2025 seasonality

7. Ranjbari, Diehl, Dalla Chiara & Goodchild: What is the Right Size for a Residential Building Parcel Locker? Transportation Research Record 2677(3), 2020 data; published 2022/2023: https://journals.sagepub.com/doi/10.1177/03611981221123807

8. Our calculation from sources 22, 2025

9. National Multifamily Housing Council / Kingsley Associates: 2018 Package Delivery Survey infographic, 2018: https://www.nmhc.org/globalassets/research--insight/package-survey-2018/nmhc_infographic.pdf

10. Pitney Bowes: Parcel Shipping Index press release (28 Mar 2023), 2022: https://www.investorrelations.pitneybowes.com/news-releases/news-release-details/pitney-bowes-parcel-shipping-index-finds-us-volumes-one-year-and

11. Pitney Bowes: Parcel Shipping Index report 2025 (US page), 2025: https://www.pitneybowes.com/us/shipping-index.html

12. Luxer One: How Granite Point Apartments Upgraded Their Package System and Won Back Resident Trust, 2026 (page ~55 days old as of Oct 2026): https://www.luxerone.com/resources/customer-stories/customer-story-how-granite-point-apartments-won-back-resident-trust/

13. Sadeghvaziri et al., Future Transportation 6(5):188 (MDPI): Survey-Weighted Analysis of Online Shopping and Home Delivery Use Among Adults with a Travel-Limiting Condition in the 2022 NHTS, 2022 (survey Jan 2022-Jan 2023; paper Sept 2026): https://doi.org/10.3390/futuretransp6050188

14. FHWA NextGen NHTS Newsletter, Summer 2024: chart 'Mean Number of Online Purchases Delivered Over the Past 30 Days per Person by Household Income (2022 Dollars)', 2017 and 2022: https://nhts.ornl.gov/assets/NextGen%20NHTS_Newsletter_Issue5_071524.pdf

15. Ha, J.: From online deliveries to travel behavior: Socioeconomic insights from 2009, 2017, and 2022, Transport Policy vol. 168, pp. 15-26, 2025 (data 2009, 2017, 2022): https://www.sciencedirect.com/science/article/abs/pii/S0967070X2500126X

16. US Census Bureau, Small Area Income and Poverty Estimates (SAIPE), 2024 state and county estimates, 2024: https://www2.census.gov/programs-surveys/saipe/datasets/2024/2024-state-and-county/est24all.txt

17. Luxer One: What's the ROI on Smart Lockers for Apartments?, 2025: https://www.luxerone.com/whats-the-roi-on-smart-lockers-for-apartments/

18. FHWA: Summary of Travel Trends: 2022 NHTS, Introduction and Appendix B, 2022: https://nhts.ornl.gov/media/2022/pub/2022_NHTS_Summary_Travel_Trends.pdf

19. NMHC Research Notes (June 2024, from CoStar data): Why Building 'Luxury' Apartments Brings Down Rent for All, 1Q 2024: https://www.nmhc.org/contentassets/b471b1d709144675bc1532bd7ddc731a/2024-june-research-notes.pdf

20. Our calculation from sources 5, 19 and 35, 2001-2023-2024

21. US Census Bureau: Quarterly Retail E-Commerce Sales, 4th Quarter 2025 (CB26-42), 2025: https://www.census.gov/retail/ecommerce.html

22. Luxer One: How to Prepare Your Community for Incoming Holiday Package Piles, 2025: https://www.luxerone.com/prepare-for-holiday-package-piles/

23. Gallup: More Ordering Groceries Online; Fewer Dining Out, 2026: https://news.gallup.com/poll/716090/ordering-groceries-online-fewer-dining.aspx

24. Dalla Chiara et al.: Do parcel lockers reduce delivery times? Evidence from the field, Transportation Research Part E 172 (Urban Freight Lab copy), 2023: https://urbanfreightlab.com/wp-content/uploads/2023/05/Do-Lockers-Reduce-Delivery-Dwell-Time-Evidence-from-Field_UrbanFreightLab.pdf

25. Ranjbari, Diehl, Dalla Chiara & Goodchild: What is the Right Size for a Residential Building Parcel Locker? Transportation Research Record 2677(3) (OSTI open-access copy), 2020: https://www.osti.gov/pages/biblio/2418037

26. Luxer One: University Package Volume Report: The 2025 Academic Year, 2025 academic year: https://www.luxerone.com/university-package-volume-report-2025-academic-year/

27. Leung et al. (2023), as cited in an arXiv preprint on multi-size parcel locker optimisation, 2023: https://arxiv.org/pdf/2607.01489

28. USPS: Parcel lockers (USPS News), 2021 (USPS notice of the July 2020 rule): https://news.usps.com/2021/04/07/parcel-lockers/

29. NMHC infographic; Fetch Package: Multifamily Package Volume is Approaching the Breaking Point, 2020: https://fetchpackage.com/multifamily-package-volume-is-approaching-the-breaking-point/

30. FreightWaves: FedEx, UPS peak season surcharges could drive shippers to competitors (quoting ShipMatrix), 2025 (history 2013, 2020): https://www.freightwaves.com/news/fedex-ups-peak-season-surcharges-could-drive-shippers-to-competitors

31. Our calculation from sources 25, 2020

32. Dalla Chiara et al.: Do parcel lockers reduce delivery times? Evidence from the field (Transportation Research Part E), 2023: https://www.sciencedirect.com/science/article/abs/pii/S1366554523000583

33. HUD User: LIHTC Data on HUD User (database through 2019), 2000-2019: https://www.huduser.gov/portal/elist/2021-sep-23.html

34. FHWA / ORNL: 2022 NHTS Data User Guide, 2001, 2009, 2017, 2022: https://nhts.ornl.gov/assets/2022/doc/2022%20NextGen%20NHTS%20User's%20Guide%20V2_PubUse.pdf

35. Harvard Joint Center for Housing Studies: Deteriorating Rental Affordability (Dec 2024), 2001-2023: https://www.jchs.harvard.edu/sites/default/files/research/files/harvard_jchs_rental_affordability_airgood-obrycki_2024.pdf