Several proposed changes to the rules of New York City’s housing lottery could make the system work more smoothly and efficiently for everyone, but a fetish to uncover what applicants “really” want is a fairness-defeating dead end.
May 4, 2026 – Apartments made available through New York City’s housing lottery system are an all-too-precious resource. 2025 was a good year and, even then, for new-construction rental lotteries where the lottery application period ended during that year, there were only about 13,400 units offered through the lottery.
The system that has been set up by the Department of Housing Preservation and Development (HPD) to handle the massive demand for affordable housing (also used by its partner agency, the New York City Housing Development Corporation, or HDC) is called Housing Connect. This is where lottery opportunities are advertised, where those interested in affordable housing can register and apply, and, behind the scenes, the system by which, among many other things, lottery numbers are assigned, and applicant information conveyed to developers.
Despite being the target of much criticism, it’s worth a moment to note that the system has been able to convey standardized information about housing opportunities and to handle millions of applications, with the biggest complaint being the pace of the process – not a hint of anything like corruption or other improper behavior on the part of HPD or HDC.
Yes, it takes too much time
The rules that govern the lottery process are set by HPD, which has received intense criticism for lottery rules which are said to contribute to delay. Two of the latest volleys are: (1) an Apr. 2026 brief from the Furman Center, traditionally a home for HPD alums and a pipeline for future HPD hires (the lead author of the brief is Vicki Been, initially HPD Commissioner and later Deputy Mayor for Housing and Economic Development in the de Blasio administration); and (2) a Mar. 2026 analysis from the New York Housing Conference, a housing policy and advocacy coalition.
Many of the proposals are sound (I have a couple to add), and I’ll get to them later in this article.
But a very large and very red warning flag must be waved vigorously to warn HPD off accepting the premise that applicant “preferences” as expressed when setting up their “profiles” in Housing Connect (or even when applying to a particular housing development) are anything like a fair way to create greater efficiency in the system. Since the lion’s share of genuine, fair-to-applicants efficiencies can be achieved without narrowing the pipeline, one has to wonder what the yearning to insert “preferences” is actually designed to do.
Preferences tell us less than meets the eye
Those who want to narrow the pipeline seem to think that too many New Yorkers are too promiscuous in applying for affordable housing, especially in terms of where they apply. Just limit yourself to the borough or neighborhoods you want.
If one were only to read the Furman brief — and contrary to the actual knowledge and decades of experience of the people who wrote and consulted on it — it would appear that there is no cognizance of the fact that locational preference for many people is a matter of what they already know in the absence of robust knowledge of alternatives.
Preferences can change when people are armed with more information. It is precisely for this reason that anyone and everyone in the fair housing world has long understood the importance of mobility counseling. Why Furman does not see this is perplexing. This service provides people with more information about neighborhoods, their amenities, relevant facilities (whether parks, hospitals, schools, proximity to transit, connection to work and pre-existing networks, and a host of other concerns that might be generated when people are introduced to something new).
New York City has done very little mobility counseling. Indeed, it is a testament to the desperate need for affordable housing that so many people apply for housing opportunities in neighborhoods they don’t know or don’t know well despite being provided little or no helpful information.
Rather than trying to restrict applications by forcing people to name boroughs or neighborhoods of preference (and then speculating that this restriction will materially improve “yield”), the city ought to be expanding its mobility counseling, including harnessing Housing Connect to do some of the work (see bottom box). Handled correctly, the processing burden at the far end of the process — even without reducing applicants at the front end — could be less than it is (and far less than complained about).
A classic example of Ivory Tower-ism
It seems straightforward to Furman that people should at least select unit types when applying for a development, rather than applying for any housing in the development that might be available. It’s more complicated.
It is often the case that someone may be eligible for more than one unit type (a studio and one-bedroom, for example; or a one-bedroom and a two-bedroom). Urging or requiring people to pick is misleading. In the first instance, Housing Connect does not currently make floor plans available as part of the information that is posted. Depending on the urgency of one’s wanting or needing to move, one might very well be open to the one-bedroom if it had a dining area that could serve as a bedroom for a young child, but uninterested if there is no such separate space. “Preferring” a two-bedroom might very well leave you without an apartment that could adequately serve your needs.
Moreover, it is sometimes the case that, by the time your lottery number is called, one or more unit types are no longer available. (Because of various preferences – the remaining community preference, municipal worker and veteran preference, etc., being closed out is a phenomenon experienced more by those who are in the general applicant pool of New York City residents without preference.) The lottery process should not urge that kind of gambling.
Debunking much of the myth of the burden of ineligible applicants
There can be cases (more with high-rent units rather than lower-rent units) where a marketing agent can go through an entire lottery list and not fill a particular unit type. But, in general, marketing agents don’t have to deal at all with the vast majority of applicants: all units are taken with tens of thousands of applicants never bearing on the process. Then there are the people who have applied even though their combination of household size and household income does not meet the requirements of any of the unit types available. As Furman admits at one point, any applicant in this category is deemed apparently ineligible and is removed from further consideration.1 Any marketing agent worth its salt (that is, even minimally competent) can take the information already provided to it by Housing Connect and, without the use of any sophisticated software tools, identify those non-subsidy applicants who are not eligible. This process is quick and easy.2
For the not inconsiderable number of applicants claiming subsidy, the process is currently more difficult because of the different rules applying to different subsidies (including differences in how much of an applicant’s income is permitted to be counted, differences in whether and how utilities are to be factored in, etc). There is a subset of subsidy applicants who turn out either: (a) to be eligible for a unit based solely on earned income; or (b) are over-income for all unit types based on earned income, but it would surely be helpful if more detailed information on how a particular subsidy affected a particular applicant for a particular development were made available.
Nevertheless, a very big dent in “ineligibles” can already be made without much effort.
Common ground: reducing the application period
Currently, the application period for smaller developments is 21 days; for larger developments the period is 60 days. Both Furman and the New York Housing Conference urge a reduction in the longer time period. They are correct.
Ask any New Yorker looking for non-lottery housing, “When you see an ad for an apartment, how fast do you think you should react?” The non-delusional answer (which is the one you’ll get most frequently) is either “today” or “in the next couple of days.”
If Housing Connect makes sure to be sending all registrants at least a round-up of new housing availabilities weekly (except for the temporary and woe-begotten “re-rental” system, Housing Connect is not first-come, first-served, so applicants have breathing space), and if HPD substantially ups its mobility counseling game so that more people are more informed about more places both before a lottery begins and while it is ongoing, the 60-day period can reasonably be shortened substantially, certainly to no more than 30 days.
Common ground: tightening deadlines and simplifying documentation
A savings of 30 days or more is not insignificant, but it is at the back end that most delay occurs. Housing Connect registrants should, to the extent that this is not already being done, get periodic reminders to update their household information, including their financial information. Such reminders should certainly go out whenever an application is completed.
Thereafter, all deadlines should be strictly adhered to. That includes the deadline for an applicant who is reached to submit required documentation, the deadline for marketing agent review, the deadline for applicant supplementation, and a deadline for marketing agent action.
And “yes” to simplifying required documentation, something that will require, among other things, more inter-agency cooperation.
The outmoded “batching” system
The procedure of having marketing agents proceed with reviewing batches of applicants was intended to preclude marketing agents from offering units out of order. But offer-order can be separated from information collection. It should not be difficult for HPD, based on historical experience, to determine how far into the general (non-preference) list and how far into the specialized preference or set-aside units (the latter for applicants with certain disabilities) the marketing agent can be expected to go.
Requests for documentation (along with a disclaimer that the marketing agent might not reach the applicant) should be sent to the number corresponding to historical experience, plus some margin.
That request for documentation should occur as soon as lottery numbers are determined and should be part of an email from Housing Connect providing the lottery number and the request for documentation. That will get the submission clock running and have documentation from many applicants available when the marketing agent is ready to review.
- 1.
Furman Brief, at 14.
- 2.
Under the current system, if Housing Connect sees, based on an applicant’s profile, that the applicant appears not to qualify for any unit type, it urges the applicant to update his or her profile information (household size and/or household income), but permits the applicant to proceed.
Furman wants that scenario to result in a mandatory bar against the applicant proceeding.
In the first instance, this would not be able to be applied to applicants claiming subsidy because the requisite information is not at hand. (These applicants, by the way, are disadvantaged by an eligibility checker that does not explicitly state that inadequate earned income is not a bar if an adequate subsidy is available.)
In the second instance, nothing is gained by a complete bar when it is so easy to perform the disqualification at the marketing-agent processing end.
It does make a difference: there are cases where appeals from apparent ineligibility have been successful.
Concurrent processing
Under current HPD rules, marketing agents, subject to agency approval, can process non-preference applicants simultaneously with preference and set-aside applicants (so long as there remains “a sufficient number of units to fulfill the preferences and so long as non-preference applicants are not processed or units that are intended for applicants eligible for set-asides”).3
Concurrent processing should be mandatory and the ability to handle this task appropriately should be a prerequisite to permitting a marketing agent to continue in the role. Concurrent processing is quicker, and it is fairer in terms of allocation of units (deferred processing of non-preference applicants means a higher rate of being closed out of particular apartment types).
It’s 2026: no more written/mailed submissions
Everyone registered with Housing Connect – either by himself or herself or with assistance – has managed to get registered with an electronic system. It is an affectation to pretend that electronic submission is too much of a burden for New Yorkers applying for affordable housing. The percentage of applications submitted by mail is currently vanishingly small, but it still slows down the process.
I am not familiar with the current process of submitting documentation, but, if not done electronically only, that in itself is a processing nightmare. Ideally, Housing Connect would be the hub for submission so that HPD (or HDC, as the case would be) would already have documentation on hand when reviewing any determinations made by marketing agents.
The auditing question and the broader problem of marketing agent variability
Suggestions have been advanced to eliminate agency review of each file and replace the process with a selective auditing function. It’s not a bad idea in principle, but it runs into two practical difficulties.
First, the moment that selective audits become the rule is the moment that there will be an inadequate number of staffers assigned to do a robust number of audits.
Second, the enormous range between and among marketing agents in terms of capacity and expertise is not sufficiently appreciated. That some marketing agents have complained about the difficulty of determining who, based on non-subsidy-applicant provided information, is apparently eligible or ineligible is a giveaway.
So is a quick gander at the temporary system for handling “re-rentals,” a system that abandoned Housing Connect for having to sort through individual property managers and marketing agents. Take a look at the variability of the quality of information provided by each; you’ll be skeptical both about the audit approach and about whether HPD is being sufficiently rigorous in qualifying marketing agents.
A citywide lottery system
Furman offers a proposal on this,4 but wait: it turns out that this is another backdoor way to introduce preferences. Furman first analogizes to the preference list that families use in the school-selection system5; it concludes with noting that HPD could continue the random approach or determine priority “taking other factors into account” like “households who have waited the longest” or “have other particular attributes.”6 I’m reminded of GOP proposals to “reform” Social Security: there the promise is a more stable system but the idée fixe is cutting benefits or requiring more years of work; here, the promise is efficiency but the idée fixe appears to be trying to create a hierarchy of which income-eligible families “deserve” affordable housing more. This zombie needs to be slain once and for all.
Substantial progress is possible without narrowing the pipeline
Keep the pipeline open, expand mobility assistance, implement concrete reforms that we know will cut substantial portions of existing delay. That’s a winning combination.
- 3.
See Marketing Handbook, 4-3(C)(6), at 23.
- 4.
See Furman Brief, at 40-44.
- 5.
See Furman Brief, at 41.
- 6.
See Furman Brief, at 44.