A Flawed Law: Reforming California’s Housing Element
By Paavo Monkkonen, Michael Manville, and Spike Friedman
In both 2017 and 2018, the state legislature passed bills seeking to reform California’s Housing Element Law. More recently, Gov. Gavin Newsom indicated he intends to reform the law further, saying that the law as currently written enables a statewide approach to housing of “neglect and denial.”
What is the Housing Element law, and why is it attracting so much attention? This issue brief introduces the law and highlights a sometimes misunderstood feature of its core planning tool: the Regional Housing Needs Assessment (RHNA) process. We focus in particular on how RHNA addresses the production of income-restricted, affordable housing. In doing so we emphasize a fundamental problem with RHNA: It creates needless tension between subsidized and market-rate housing, and as a result generates too little of either.
We suggest a solution that separates the goal of having cities carry a “fair share” of affordable housing from the goal of building enough housing to accommodate future growth. Specifically, we think the state should simplify its fair share requirement, and mandate that every city carry an equal percentage of income-restricted housing and make space for an equal rate of new housing growth.
- RHNA numbers are not true “housing needs” — they are the result of a political process in which affluent cities lobby to keep their numbers low.
- The RHNA process asks cities to plan for more than 10 times the amount of subsidized housing that can be funded.
- RHNA bases “housing needs” on projections of population, rather than prices. As a result, it steers housing production away from places where housing is most expensive, and new development most necessary.
- RHNA needlessly pits market-rate and affordable housing against one another, when both are necessary to advance affordability.
- RHNA can be improved and simplified, and some northeastern states offer a model for reform. Rather than engage in a complicated process of prediction and allocation, California should simply require that all cities build a “fair share” of affordable housing.
The Housing Element Law has two goals. It primarily seeks to ensure that cities zone for enough new housing to accommodate future population growth. But it also tries to ensure that some of that zoned capacity for housing is specifically for income-restricted, subsidized affordable housing. The way it does so, however, is convoluted and — until recently — largely toothless.
The RHNA process has three stages. First, the state Department of Finance and regional Councils of Government, in consultation with the Department of Housing and Community Development, project population growth and household formation for each region in the state. Then, each region, through its metropolitan planning organization, allocates a portion of this projected housing “need” to its constituent cities. This process involves not just assigning projected units to each city, but also subdividing those units by income groups (e.g., each city will “need” a certain number of low-, middle- and higher-income housing units). These final figures are the cities’ RHNA numbers. Lastly, the cities must then revise the Housing Elements of their General Plans to demonstrate they have more than enough zoned capacity to accommodate this number of housing units.
This process is problematic for at least two reasons. First, little thought is given to how these needed units — especially the affordable units — might be built. To say that there isn’t enough public subsidy to build all the income-restricted housing that the state deems necessary is a wild understatement. The fifth RHNA cycle, covering 2013–2021, estimated that California would need over 665,000 income-restricted units, or roughly 80,000 affordable units per year. A 2015 report from the Legislative Analyst’s Office noted that the Low Income Housing Tax Credit (LIHTC) program, the primary subsidy for new income-restricted housing, finances the production of about 7,000 units in California annually. Meeting our estimated need for affordable housing would require annual expenditures in the low tens of billions of dollars.
The second problem is in some ways a solution to the first problem, albeit not a good one. RHNA only requires that cities demonstrate, in their plans, that they have space that can potentially hold the needed income-restricted housing. But nothing in the law ensures that any income-restricted housing will actually be built, nor that these potential sites of income-restricted housing must actually be reserved for that purpose. Municipalities uninterested in building affordable units are thus under no legal obligation to do so. Although they must show a zoned capacity for this housing in their Housing Element, if anyone actually proposes such housing the cities can simply block it through their usual development review process. This aspect of the law is no secret. On its website, Palos Verdes Estates, a small, wealthy city in the South Bay of Los Angeles County, explains to residents that the state only requires potential space, not units (Figure 1).
Figure 1. Excerpt from Palos Verdes Estates Online Housing Element Questions & Answers document
The predictable result of this loophole is that many cities don’t build income-restricted housing. The few that do end up carrying a disproportionate share of the state’s affordable housing, which exacerbates spatial inequality within metropolitan areas. Part of SB35, a bill passed in 2017, involved tracking whether cities built a “sufficient share” of the income-restricted housing they were assigned during the RHNA process. Only 24 jurisdictions out of the state’s 540 met RHNA goals.
Another way to see the unequal distribution of income-restricted housing is to look at where we build LIHTC units. We reviewed the production of LIHTC housing in Los Angeles County’s 88 cities: Thirty have constructed no LIHTC units ever, while 24 others have built only one or two buildings. The city of Los Angeles, which is home to roughly 40 percent of the county population, has 70 percent of the county’s 1,235 income-restricted buildings built through the LIHTC program.
To our knowledge, Palos Verdes Estates has zero income-restricted housing units. But it is in full compliance with RHNA. It has identified a site that could hold income-restricted housing, and that’s all that matters.
Can RHNA really identify housing “needs”?
Behind RHNA’s income-restricted housing loophole lurks a larger, conceptual problem. The RHNA process predicts how much housing a city “needs,” and then breaks that need down by price — so for example the 2013–2021 RHNA numbers for Los Angeles said the city would need 46,000 units for people earning less than 120 percent of the area’s median income, and 32,000 units for people earning above that.
Two problems afflict this exercise. First is that the RHNA numbers are not true “housing needs” — they are the results of a political process where cities lobby and, often, fight to keep their numbers low. That’s why Beverly Hills, which is no one’s idea of affordable, was famously determined to “need” only three new units of income-restricted housing between 2013 and 2021, while other similarly sized cities and much more affordable cities were determined to “need” hundreds of units.
More generally, even if cities did not lobby, the entire exercise of predicting how much housing California will need in the next eight years is a little odd. In general, the market for a good is working well if anyone who wants the good can find it, and people who sell the good make enough money to keep them in the business (but not much more). For most markets, arriving at this point doesn’t require complicated prediction. No one knows, for example, how many eggs California will need to produce in the next week, or the next decade, for everyone who wants an omelette to enjoy it at breakfast. We “solve” this intractable egg problem by allowing farmers to respond easily to the demand for eggs. When egg prices rise, no one stops farmers from getting more eggs to market. The same isn’t true for housing. Prices rise, but housing is hard to build in expensive areas, and often hard to build, period. Is housing more complicated than egg production? Of course. But the basic point holds: California’s problem is not its trouble predicting the “right” numbers, but its myriad rules preventing housing from being built in places with clear demand for it.
This point was not always obscure. RHNA was originally written, in part, to push back on the many restrictions cities had placed on housing production. In practice, however, RHNA became something different. By framing housing needs in terms of population and household growth, rather than prices in the existing market, RHNA steers policymakers away from the logically obvious but politically inconvenient point that housing production should be concentrated in places where housing is most expensive. Previous RHNA cycles, in fact, have arrived at the exact opposite conclusion — that more housing is needed in places where prices are lower.
RHNA generates this mistaken conclusion because it looks at places that have grown slowly (because they restrict housing production) and then projects that growth will be slow there in the future. This predicted slow growth, in turn, leads to the conclusion that these cities “need” less housing. Slow growth thus becomes a self-fulfilling prophecy. The process rewards rather than sanctions cities that refuse to build, and allocates housing “need” to places where housing is needed least.
This flawed needs assessment creates a final problem. In subdividing housing allocations by income, the process creates the mistaken impression that some places have “enough” of certain kinds of housing. When combined with the requirement that cities identify specific sites for low-income housing, this impression can make market-rate and subsidized housing appear to be at odds with each other.
For example, in its most recent Housing Element, the city of Los Angeles listed a large parcel in Hollywood as a site that could hold over 100 units of affordable housing. In 2019, however, the city approved a 950-unit mixed-use project on the parcel, which included no income-restricted units. The project angered some activists, and did so for a number of reasons, but one prominent objection was that Los Angeles is on track to meet its RHNA goals for market-rate housing. It thus seemed like the city was sacrificing needed affordable units for “unnecessary” market-rate units.
That logic, however, is misguided. The fact that Los Angeles might hit its RHNA target for market-rate housing should in no way suggest the city has “enough” market-rate housing — a simple look at the price of Los Angeles’ market-rate housing should tell us the city needs much more of it. Similarly, the fact that Los Angeles has zoned relatively few places as suitable for low-income housing should not tell us those places must be protected at all costs — it should tell us that Los Angeles needs to change its zoning. RHNA creates conditions where people fight over the few and shrinking sites where multi-family housing is allowed, when the real goal should be to create many more of those sites. Zoning is not immutable, and cities can change it with the stroke of a pen. All that is needed is political will.
Pitting market-rate and affordable housing against one another is inaccurate and counterproductive. Both are necessary, and both advance the larger goal of affordability. Saying this does not make the two types of housing equivalent. It is true that income-restricted housing is often the only sort of new housing that can directly assist very low-income people, and for that reason income-restricted housing is essential. But many low- and moderate-income people live in older market-rate housing, and they face rising rents when market-rate housing is in short supply. More market-rate housing, especially if it is built in expensive areas where higher-income people want to live, eases the pressure these households face. We cannot solely build market-rate housing. But failing to build it does the disadvantaged no favors.
Two simplifying reforms
RHNA can be improved. The problems we describe stem from the law being overly complex, both in its efforts to determine how much and where different kinds of housing should be built. Not all places have gone down California’s cumbersome path, and these other places offer examples we can learn from. A number of northeastern states, for instance, have enacted laws to override exclusionary zoning, and their model appears to have been more effective than California’s. Massachusetts State Law 40B, for example, has led to increased production of rental housing in job-rich areas that placed strict limits on multi-family housing. Similarly, legislative action bolstered by strong judicial intervention in Pennsylvania led to the development of more multi-family housing in suburban areas than in neighboring states.
One feature of the northeastern model is the way states determine whether cities are in compliance. These states do not use convoluted predictions, or care about potential sites for housing. Instead, a city is out of compliance if less than 10 percent of their housing stock is income-restricted. And if a city is out of compliance developers can bypass many local reviews when building low-income housing.
This approach has two straightforward benefits. First, it is simple. There is no need for a complex modeling effort to determine “needs” every eight years. More importantly, it avoids politics. Without a complex allocation system, there is no one to lobby and nothing to game. Every city should have at least 10 percent affordable housing. This number might strike many as too low, but we should walk before we run. In California, 10 percent would represent drastic improvement. In the last RHNA cycle, only one in 10 California cities were asked to zone for more than 10 percent of their stock as income-restricted housing. Eventually we could move towards the French goal of 20 percent subsidized housing in every municipality.
The simplicity of this approach, and the benefits of removing housing targets from the local political process, may also be usefully applied to RHNA overall. Rather than asking Councils of Government to develop a methodology and negotiate the allocation of market-rate growth in each city, the state could simply require that all cities zone for 10 percent more housing than they currently have. During the most recent RHNA cycle, the median city’s total RHNA number was roughly 5 percent of its housing stock. Only a quarter of California cities were asked to zone for growth of 10 percent or more in housing stock. Moreover, prior RHNA cycles have allocated more housing needs to cities with lower-income residents located far from job centers. A simple proportional allocation of RHNA numbers across cities could improve equity between cities and between people, and help the state meet its goals for infill development. Eventually, perhaps, we could move to a model actually based on demand, where cities with higher rents and prices be required to zone for more growth than others.
California’s state government has long recognized that many local governments, left to their own devices, will not build enough housing. The state has responded with a system designed to preempt municipal regulation and increase both the supply of housing generally, and of subsidized housing in particular.
But that system has several flaws. The RHNA process asks cities to zone for much more subsidized housing than we have subsidies available to build, and cities face few consequences when housing they zone for goes unbuilt. Yet no one can live in zoned capacity — the state needs housing, not just planning.
RHNA’s distribution of expected construction, moreover, is highly inequitable across cities. We ask poorer cities farther from jobs to build more subsidized housing than high-amenity cities where demand is high.
The process also pits affordable housing that we sorely need against market-rate housing that we also sorely need, on the limited number of multi-family sites in municipal housing elements.
Finally, the state frames housing in terms of “needs” that ignore market signals. This can generate the obviously incorrect conclusion that wildly expensive cities have built “enough” housing. We will know the state has “enough” housing when people who want it can find it, without carrying crushing cost burdens. That is the goal RHNA should be oriented around.
Paavo Monkkonen is the Lewis Center senior fellow for housing policy and an associate professor of urban planning at the UCLA Luskin School of Public Affairs. Michael Manville is an associate professor of urban planning at UCLA Luskin. Spike Friedman is a Master of Urban and Regional Planning student at UCLA Luskin.
 It also estimated the state needed 450,000 market-rate units during this period.
 The definition of a suitable site for low income housing is simply any residential parcel zoned for density of 20 units per acre (Government Code Section 65583.2(c)(3)(B)(iii) and (iv)).
 There are other ways to create income-restricted housing, such as inclusionary housing ordinances, but a lack of comprehensive data on these makes a comparison across cities complicated.
 It does, however, receive ample federal housing assistance. Together with the neighboring small city of Rolling Hills Estates, it received almost $120 million# in benefits from the Mortgage Interest Deduction in 2016.