Local tax revenue reflects a community’s shared effort to support vital public services that all Californians need to live in our cities and counties, such as education for students in K-12 schools and community colleges, housing, health care, public parks, and libraries. When tax breaks provide advantages to some taxpayers over others, it not only creates inequities but can also lead to revenue losses that compromise the ability of schools and local communities to provide essential services for Californians. This is the case with commercial and industrial property taxes across California, and why voterswill be asked in fall 2020 to vote on Proposition 15, an amendment to the state Constitution that would change how commercial and industrial properties are taxed to provide more revenue for schools and communities.
Under Prop. 15, commercial and industrial properties would be taxed based on their market value rather than their purchase price. By moving from a property tax system based on purchase value to one based on market value, Prop. 15 would raise an estimated $6.5 billion to $11.5 billion annually in property tax revenues for K-12 schools and community colleges, counties, cities, and special districts, according to the Legislative Analyst’s Office.
FAQ: Understanding Commercial Property Tax & Revenue in California
How Are Commercial and Industrial Properties Taxed Today? The general property tax rate for California commercial and industrial properties has been capped at 1% of assessed value since voters approved Prop. 13 in 1978. Counties determine the assessed value of commercial and industrial properties based on the property’s purchase price plus an annual adjustment for inflation not to exceed 2%. Counties collect property taxes and are generally only allowed to reassess properties to their market value when they undergo a change in ownership or new construction.
How Is Revenue From Commercial and Industrial Property Taxes Distributed Across California? Revenue received from the taxes paid by commercial and industrial property owners is distributed to counties, cities, K-12 schools and community colleges, and special districts (such as public utility districts and fire protection districts) for services provided to Californians, based on complex state laws. The share of countywide property tax revenue going to each local entity is largely based on the distribution of these revenues dating back to the mid-1970s – before Prop. 13 was enacted and each local entity was able to set its own property tax rates. This means that there is wide variation among counties in the share of revenue going to – and the level of services provided by – each type of local government.
Why Are Commercial and Industrial Property Taxes Inequitable for Californians and in Need of Reform? The property assessment limits set by Prop. 13 mean that an owner that purchased a commercial or industrial property several decades ago pays far lower taxes than an owner that recently purchased a similar property – leading to inequity among local businesses and a significant loss of revenue at the expense of schools and local community services. Schools and local communities are losing significant revenues every year as properties that have not changed ownership in many years are assessed at values much lower than their market values. Additionally, when a property changes hands, commercial and industrial property owners can more easily avoid reassessment than residential property owners due to the laws defining ownership changes and the complexity of business property ownership.
Report: Raising Revenue for Schools and Local Communities, Changing California’s Inequitable Taxing of Commercial Properties, and Understanding Proposition 15
Local tax revenue reflects a community’s shared effort to support vital public services that all Californians need to thrive in our cities and counties. This ranges from education for students in K-12 schools and community colleges to access to housing, health care, public parks, and libraries. These vital public services are supported by tax revenues from commercial and industrial properties – many of which are still taxed based on purchase prices that are more than four decades old. California voters will be asked in fall 2020 to vote on a measure known as Proposition 15, an amendment to the state Constitution that would change how commercial and industrial properties are taxed and provide more revenue for schools and local communities to support services Californians rely upon.
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
Across California and the United States, the push for bail reform has gained momentum with increasing awareness and research showing the disproportionate impact the money bail system has on people of color and low-income households. In California, it’s estimated over two-thirds of people detained in jails – around 47,000 – have not been sentenced for a crime, a number that includes both those who cannot afford bail and those who are awaiting sentencing post-conviction. Los Angeles County alone is the largest jail system in the US and houses over 1 in 5 of adults who have not been sentenced for an alleged crime in California.
Enter Proposition 25 that will appear on the November 3, 2020 statewide ballot and asks California voters to decide whether a 2018 state law that effectively ends money bail should take effect. If voters approve Prop. 25, judges will be able to utilize risk-based assessment tools – examining population links between rearrest or reconviction and individual factors such as age, gender, or criminal record – to determine if individuals detained for certain crimes can be released before a court appearance rather than posting money bail.
Read the full report that discusses racial, economic, and gender disparities embedded in the money bail system and why efforts to reform California’s bail system also aim to address the wide racial disparities seen in the criminal justice system.
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
Over many years, California lawmakers and voters adopted a series of harsh, one-size-fits-all sentencing laws that prioritized punishment over rehabilitation, led to severe overcrowding in state prisons, and disproportionately impacted Black and Latinx Californians – consequences that many families still feel today. California began reconsidering its “tough on crime” approach a little over a decade ago as prison overcrowding reached crisis proportions and the state faced lawsuits filed on behalf of incarcerated adults. Ultimately, a federal court in 2009 ordered California to reduce overcrowding to no more than 137.5% of the prison system’s capacity – an order that remains in effect today.1
State-level reforms – enacted into law through voter approval of ballot propositions as well as through legislative action – have focused on reducing incarceration, promoting more effective pathways to rehabilitation, and addressing the disparate impacts of criminal justice policies on people of color, particularly Black and Latinx communities.2 With these reforms, both the prison population and crime rates are down substantially, showing that California’s efforts to reduce mass incarceration, while far from complete, are working.
1. The Number of Adults Incarcerated by the State Has Declined Substantially
Adults who are incarcerated by the state fall under the jurisdiction of the California Department of Corrections and Rehabilitation (CDCR). Most of these adults – more than 9 in 10 – are held in 34 state-owned prisons. The remaining men and women are housed in other locations throughout California, including in public and private facilities under contract with CDCR. (Until recently, thousands of state prisoners were housed in out-of-state facilities.) The number of adults incarcerated by the state exceeded 173,000 in 2007, when state prisons were crowded to roughly double their capacity. By June 2020 – following years of criminal justice reforms and, more recently, new policies adopted in response to the COVID-19 pandemic – the number of adults incarcerated had dropped by more than one-third, to 113,403.3
2. With Declining Incarceration, California Ended the Use of Out-of-State Prisons and Has the Opportunity to Begin Closing State-Owned Prisons
In 2006, California began transferring incarcerated adults to facilities in other states to help reduce overcrowding in state prisons. State policymakers adopted this approach even though “out-of-state private prisons create significant barriers to rehabilitation and humane conditions of care.”4 The number of Californians moved out of state peaked at more than 10,000 in the early 2010s. However, as incarceration declined, California was able to gradually reduce its reliance on facilities in other states before finally terminating its last out-of-state contract in June 2019. Moreover, due to the ongoing decline in the prison population, California is on track to soon end the use of all in-state contract facilities for men, and the recently enacted 2020-21 budget package envisions closing two state-owned prisons in the coming years.5
3. California’s Incarceration Rate for Adults Has Fallen to a Level Last Seen in the Very Early 1990s
As California implemented criminal justice reforms, the incarceration rate – the number of adults incarcerated by the state for every 100,000 residents – has plummeted. California’s state-level incarceration rate dropped to 314 per 100,000 in June 2019, down by more than one-third (34%) from the recent peak of 476 per 100,000 in June 2006. Moreover, the June 2019 incarceration rate was slightly below the June 1990 level of 315. Still, incarceration of California men and women – disproportionately Black and Latinx Californians – remains high compared to earlier years. For example, in the late 1970s the state incarcerated fewer than 100 people for every 100,000 residents.
4. Meanwhile, California’s Property and Violent Crime Rates Remain at Historic Lows
California’s property crime rate – the number of property crimes per 100,000 residents – was 2,290 per 100,000 in 2019, far below the peak of 6,881 in 1980. The violent crime rate was 434 per 100,000 in 2019, less than half the peak of 1,104 in 1992. This latter rate is up modestly compared to the low of 393 per 100,000 in 2014, partly due to technical factors.6 Nonetheless, California’s violent crime rate resumed its decline after 2017 and is now below the 1969 rate (449 per 100,000).
5. California’s Incarceration Rate and Crime Rates Are All Down Substantially Since the Mid-2000s
Incarceration and crime rates are all down substantially compared to their levels in 2006 – shortly before state policymakers and the voters began enacting reforms to California’s criminal justice system. As noted above, the incarceration rate of adults dropped by more than one-third (34%) from June 2006 – the recent peak – to June 2019. During approximately the same period (2006 to 2019), California’s property crime rate fell by more than one-quarter (28%) and its violent crime rate declined by nearly one-fifth (19%). These statistics contradict the common, yet unsubstantiated, claim that reducing mass incarceration will cause crime rates to spike. In fact, California’s experience and a large body of research highlight the weak link between incarceration and crime.7
Conclusion
California’s experience with criminal justice reform provides further evidence that reducing mass incarceration of men and women can go hand-in-hand with lower crime rates. But California’s work is not done – policymakers can and should do more to decrease incarceration of Californians, particularly given the impact of the COVID-19 pandemic on incarcerated adults, prison staff, families, and surrounding communities. This is especially important for Black and Latinx men and women and their families, who are bearing the greatest burdens of COVID-19 in prisons and the broader community. In recent weeks, the state has been moving in the right direction: The prison population has fallen more rapidly than anticipated due to early releases and other policies advanced by Governor Newsom to slow the spread of the coronavirus behind prison walls. With these steps, the Governor has created an opportunity to plan for the closure of several state-owned prisons over the next several years – assuming, at a minimum, that current criminal justice reforms remain in place and that the state prison population does not again begin to rise.
Downsizing California’s costly prison infrastructure would allow the state to reduce the size of the corrections footprint on the state budget. This, in turn, would free up resources that could be used for reentry assistance and other services that can help to promote rehabilitation, reduce poverty, and strengthen families and communities – particularly Black and Latinx communities, which have been disproportionately impacted by the pandemic and the deep recession that it triggered as well as by generations of discrimination at the hands of the criminal justice system.
1 The US Supreme Court upheld this order in 2011. California has been in compliance with this order since February 2015. See California Department of Corrections and Rehabilitation, Three-Judge Court Monthly Update (July 2020).
2 For an overview of key state-level reforms, see Scott Graves, State Corrections in the Wake of California’s Criminal Justice Reforms: Much Progress, More Work to Do (California Budget & Policy Center: October 2018), pp. 4-10.
3 CDCR’s in-custody population further declined to 103,169 as of August 12, 2020. This substantial ongoing decline reflects early releases and other policies adopted by the Newsom Administration that are intended to slow the spread of the coronavirus in state prisons.
4 Randall G. Shelden and Selena Teji, Collateral Consequences of Interstate Transfer of Prisoners (Center on Criminal and Juvenile Justice: July 2012), p. 4.
5Department of Finance, California State Budget 2020-21 (June 2020), p. 80.
6 Technical factors “related to crime classification and reporting” contributed to the slight – yet temporary – uptick in violent crime rates after 2014. Mia Bird, et al., The Impact of Proposition 47 on Crime and Recidivism (Public Policy Institute of California: June 2018).
7 Don Stemen, The Prison Paradox: More Incarceration Will Not Make Us Safer (Vera Institute of Justice: July 2017).
You may also be interested in the following resources:
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
Introduction
Prop. 2, which will appear on the November 4, 2014 statewide ballot, would amend the state Constitution to substantially revise the rules for the state’s existing rainy day fund, require state policymakers to pay down certain state liabilities each year for 15 years, and create a new state budget reserve for K-12 schools and community colleges. This Issue Brief provides an overview of this ballot measure and the policy issues it raises. The California Budget & Policy Center neither supports nor opposes Prop. 2. [Note: California voters overwhelmingly approved Prop. 2, with 69.1% supporting the measure and 30.9% opposed.]
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
What Would Proposition 1 Do?
Prop. 1 would allow California to sell $4 billion in bonds to support affordable housing development and to help veterans and low- and moderate-income homebuyers purchase homes. Prop. 1 was placed on the ballot through legislation enacted last year, the Veterans and Affordable Housing Bond Act of 2018 (Senate Bill 3, Beall), which was part of the broader legislative housing package passed by state policymakers to help address California’s housing affordability crisis. Of the $4 billion in bonds sold under Prop. 1, $3 billion would be repaid over several decades, with interest, from the state General Fund, and $1 billion would be fully repaid by veterans receiving bond-funded home loans.
What Problem Does Proposition 1 Aim to Address?
California faces a housing affordability crisis. Among all Californians in 2016, more than 4 in 10 households (42.1%) across the state paid more than 30% of their incomes toward housing costs, a level considered unaffordable according to the US Department of Housing and Urban Development. More than 1 in 5 (21.0%) were severely housing cost-burdened, spending over half their incomes on housing, according to a Budget Center analysis of US Census Bureau data.
Californians with low incomes have particularly high rates of housing cost burden. Among households with incomes below 200% of the official federal poverty threshold, more than 1 in 2 (57.3%) paid more than half of their income toward housing costs in 2016. Unaffordable housing burdens affect residents in all parts of the state, even in areas where housing costs are lower, because incomes in those areas tend to be lower as well. Moreover, California’s high housing costs are a key driver of the state’s high poverty rate. Under the Supplemental Poverty Measure, an improved poverty measure that accounts for differences in local costs of living, nearly 1 in 5 Californians (19.0%) were living in poverty from 2015 to 2017 — giving California one of the highest poverty rates in the country, statistically tied for first (with Florida and Louisiana) among the 50 states.
When individuals do not have access to housing they can afford, the consequences can be serious. This is because safe, stable housing is a key foundation for short-term and long-term health and well-being. Unaffordable housing costs can force households into substandard housing, which is linked to poor health outcomes. Lack of affordable housing can also force families to move more often, and this housing instability is linked to negative health outcomes for both adults and children, as well as worse educational outcomes for children. In the most extreme cases, high housing costs can push households into homelessness, with particularly negative effects on children’s physical and mental health. Addressing the problem of housing affordability can help prevent this cascade of negative health, behavioral, and educational outcomes. Improvements in individual health and well-being can also translate into broader public benefits, as healthier individuals are less likely to need public services and supports to address health or mental health needs.
A key driver of California’s housing affordability crisis is an inadequate supply of housing. The Department of Housing and Community Development estimates that 180,000 new housing units need to be built each year to keep up with housing demand, but over the past 10 years only 80,000 units have been built per year on average. Given the large scale of this problem, an array of different policy strategies are needed to push California’s housing supply to catch up with the need for housing for current and future residents.
One strategy is direct state investment in building affordable housing and helping individual Californians afford housing. An increase in the supply of housing that is affordable to the lowest-income residents is especially needed in California, because these households are least able to afford California’s escalating rents, which have been growing faster than median annual earnings. The National Low Income Housing Coalition estimates that California has only 67 affordable housing units available for every 100 low-income households. Research in the San Francisco Bay Area has also shown that in the face of gentrification pressures, an increase in subsidized affordable housing units has nearly twice the impact in reducing displacement of low-income households at the regional level, compared to a similar increase in market-rate housing units.
What Is the Expected Impact of Proposition 1?
The bond funds from Prop. 1 would be allocated to a number of established state housing programs. Of the $4 billion total, $3 billion would support the development of affordable housing and homebuyer assistance for low- and moderate-income Californians, specifically:
$300 million to the Infill Infrastructure Grant Program, for new construction and rehabilitation of infrastructure that supports high-density affordable and mixed-income housing in infill locations
$300 million to the CalHome Program of the Self-Help Housing Fund,* for forgivable loans for projects that create multiple affordable homeownership units, self-help mortgage assistance, and mobile homes
$300 million to the Joe Serna Farmworker Housing Grant Fund,* for construction, rehabilitation, and acquisition of housing for agricultural employees and their families
The remaining bond funds consist of $1 billion for the Cal-Vet Home Loan Program to provide homebuyer loans to veterans, without regard to income (which would be paid back to the state by the loan recipients). Altogether, the Prop. 1 bond funds would support a wide variety of housing activities, including multifamily affordable housing development, infrastructure to facilitate infill and transit-oriented housing, affordable homeownership opportunities, farmworker housing, self-help (e.g. Habitat for Humanity) and mobile home housing, and mortgages for veterans.
In the case of housing and/or infrastructure development, funds from Prop. 1 would typically make up only a portion of the total funds for any given project. Bond funds would be used by developers to leverage additional federal, state, and local public funds, as well as private financing, to cover the full costs of these types of projects, which would multiply the impact of the Prop. 1 funds.
Altogether, the Legislative Analyst’s Office (LAO) estimates that Prop. 1 would “provide annual subsidies for up to 30,000 multifamily and 7,500 farmworker households” as well as “down payment assistance to about 15,000 homebuyers and home loans to about 3,000 veterans.” The California Housing Partnership Corporation and Northern California Carpenters Regional Council, which support Prop. 1, have estimated a somewhat larger number of housing units — “nearly 50,000 new and rehabilitated housing units” — would be created with the support of the $3 billion in Prop. 1 housing funds (separate from the $1 billion for mortgage assistance for veterans). This estimate was calculated using the IMPLAN economic impact model and was based on historical per-unit costs and leveraged funding, adjusted by region, for each of the state housing programs to which Prop. 1 funds would be allocated.**
The housing produced through Prop. 1 would not be sufficient to close California’s affordable housing shortfall. However, no single policy could be expected to fully address the state’s tremendous housing affordability challenges. The bond funds would support the production of housing units affordable to low-income residents, which are unlikely to be produced by the private for-profit housing market and which would help the residents who struggle most to afford the state’s high housing costs.
What Are the Tradeoffs in Using Bond Dollars to Support Affordable Housing Through Proposition 1?
Of the total $4 billion in bonds that Prop. 1 would authorize, the $1 billion for mortgages for veterans through the Cal-Vet Home Loan Program would be expected to be fully repaid over time by the veterans receiving home loans, at no direct cost to the state, according to the LAO. However, the remaining $3 billion in bonds for other housing activities would have to be repaid over time, with interest, from the state’s General Fund. The LAO estimates that the cost to repay the bonds would be approximately $170 million each year for 35 years. Over this period, the state would pay $3 billion to repay the principal on the bonds and an estimated $2.9 billion for the interest, according to LAO estimates, for a total of $5.9 billion, or nearly twice the amount of the original bond issuance.
The General Fund dollars used to repay the principal and interest on the Prop. 1 bonds would not be available to support other public systems and supports, including those that directly support low- and moderate-income Californians — a tradeoff that should be considered when evaluating the potential impact of Prop. 1. On the other hand, as noted above, safe, affordable, stable housing is a foundation for long-term physical and mental health and is linked to improved individual educational and economic outcomes. As a result, an investment in affordable housing could contribute to healthier, more productive residents who may have less long-term need for publicly-funded health, mental health, and safety net services.
What Do Proponents Argue?
Proponents of Prop. 1 include veterans’ organizations, affordable housing advocates, business and labor leaders, cities, and environmental groups. Among these are Dignity Health, Habitat for Humanity California, United Ways of California, California Association of Veterans Services Agencies, California Housing Consortium, Housing California, Silicon Valley Leadership Group, State Building and Construction Trades Council of California, California League of Conservation Voters, and the California League of Cities. Proponents argue that Prop. 1 “directly addresses the shortage of housing by building more affordable homes — without raising taxes.” They state that Prop. 1 will benefit “hardworking people like nursing aides, grocery clerks, and teaching assistants, so they can live in the communities where they work” and “will address rising homelessness in our neighborhoods.”
What Do Opponents Argue?
Opponents of Prop. 1 argue that the bond funds “would go to a variety of programs that may or may not repay money for revolving use” and would help only “a very limited number of persons.”
Conclusion
To address the root cause of California’s housing affordability crisis, the state needs to increase its supply of housing. Increasing the supply of affordable homes in particular is especially needed, because low-income Californians are most likely to have high housing cost burdens and are least likely to be able to afford rising rents. Prop. 1 would not in itself eliminate California’s affordable housing gap, but it would take a step toward addressing the shortfall by allowing the state to sell $4 billion in bonds and use the proceeds to develop affordable housing and help veterans and low- and moderate-income Californians purchase homes.
These bonds would be paid off over 35 years, at about $170 million per year. The $1 billion in bonds to help veterans buy homes would be fully repaid by home loan recipients, at no direct cost to the state, but the $3 billion in bonds for other housing activities would incur principal and interest costs, which would be paid out of the state General Fund. These dollars would not then be available to pay for other public systems and costs supported by General Fund dollars, like pensions for public employees, health care for low-income individuals, or the criminal justice system. A key question for voters is whether developing tens of thousands of affordable housing units and assisting veterans and lower-income residents to purchase homes is worth the tradeoff of dedicating these General Fund dollars to bond principal and interest payments over time. In evaluating this tradeoff, considerations include the fact that bond funds would leverage significant federal, local, and private dollars to support housing production and that affordable housing has cascading positive effects on health and well-being, which may reduce demands for other public supports over the long term.
* Up to $360,000 of the total combined funds allocated to these programs could be used by the Department of Housing and Community Development to provide related technical assistance to cities and counties.
** This estimate did not account for the $150 million allocated to homebuyer assistance through CalHFA rather than to housing production, which would slightly decrease the estimated number of housing units produced.
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
The Budget Center hosted a half-day event in Los Angeles at the Joan Palevsky Center for the Future of Los Angeles on five important ballot measures: Prop. 1 (affordable housing bond), Prop. 2 (mental health dollars for homelessness), Prop. 5 (expanding local property tax breaks), Prop. 6 (curtailing transportation funding), and Prop. 10 (rent control).
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
Proposition 10, which will appear on the November 6, 2018 statewide ballot, would repeal a state law that limits the scope of local rent control policies that cities and other local jurisdictions are allowed to adopt. Prop. 10 was placed on the ballot by petition signatures and qualified for the ballot with key financial support from Michael Weinstein of the AIDS Healthcare Foundation. This post provides an overview of Prop. 10, discusses its expected impact, and examines other issues the measure raises in order to help voters reach an informed decision.
What Would Proposition 10 Do?
Prop. 10 would repeal a state law, the Costa-Hawkins Rental Housing Act (or “Costa-Hawkins”), that currently places restrictions on the rent control policies that cities and other local jurisdictions may choose to apply to rental housing in order to limit allowed rent increases. Costa-Hawkins currently prohibits cities from limiting rent increases in certain types of rental homes, including single-family homes, apartments that were built after February 1995 (or after an earlier date in some cities), and vacant apartments that are turning over to new tenants. By repealing Costa-Hawkins, Prop. 10 would allow cities to choose to limit rent increases in these types of rental homes in addition to the older, occupied rental apartments to which local jurisdictions may currently apply rent control if they choose.[1]
What Problem Does Proposition 10 Aim to Address?
Housing costs in many parts of California are very high, and housing is unaffordable for many residents in all parts of the state.[2][3] Many Californians now pay more than 30% of their income for housing, which the US Department of Housing and Urban Development defines as an unaffordable housing cost burden. Among all Californians in 2016, more than 4 in 10 households across the state were housing cost-burdened, and more than 1 in 5 were severely cost-burdened, spending over half their incomes on housing, according to a Budget Center analysis of US Census Bureau data. Unaffordable housing is a problem that disproportionately affects low-income households, and the majority of individuals with high housing cost burdens in California are people of color.[4] Moreover, California renters are particularly affected by unaffordable housing cost burdens. More than half of California renter households paid over 30% of their income toward housing in 2016, and nearly 1 in 3 paid more than half of their incomes toward housing (see Figure 1).
Figure 1
These high housing cost burdens have serious consequences for California families. By making it harder for families to make ends meet, unaffordable housing costs may force families to double-up to save on rent or settle for substandard housing in unhealthy, low-opportunity neighborhoods. Families who are paying more than they can afford for housing may have to cut costs by choosing low-quality child care or letting health problems go unaddressed, and they may be unable to set aside savings for emergencies or for retirement. In the worst cases, they may be pushed into homelessness. These kinds of hardships have both immediate and long-term consequences, for adults and especially for children.[5]
California’s unaffordable housing costs are a problem not simply because housing costs are high, but because they have been growing quickly relative to incomes. While median annual earnings for full-time workers in California increased 4% from 2006 to 2016, median household rents increased by 13%, more than three times as much as earnings did, during the same period (see Figure 2).
Figure 2
With earnings outpaced by living costs, families and single individuals struggle to cover the basic costs of housing, food, child care, and other needs. As a response to California’s housing pressures, rent control focuses on protecting renters from rapidly rising housing costs. Prop. 10 would allow cities to choose to expand local rent controls by repealing the restrictions on rent control imposed by Costa-Hawkins.
What Is the History of the Costa-Hawkins Rental Housing Act?
In the 1970s and 1980s, an earlier period of rapidly rising rents, several California cities adopted local policies that limited the rent increases that landlords were allowed to impose, policies known as rent control or rent stabilization. Landlords and housing developers sought to restrict the allowed scope of these types of local policies through a failed state ballot initiative and repeated failed legislative proposals in the 1970s, 1980s, and early 1990s. Finally, in 1995, the Costa-Hawkins Rental Housing Act, sponsored by the California real estate industry, passed the Legislature and was signed by Governor Pete Wilson.[6]
Through Costa-Hawkins, the state imposed three limits on the scope of local rent control laws.[7] First, single-family homes were exempted from rent control. Secondly, rent control could not apply to any newly built housing starting February 1, 1995. For cities that already had some form of rent control, Costa-Hawkins back-dated this second restriction on rent controls on new housing to the date of the local ordinance. (For example, since Los Angeles instituted rent control in 1978, housing built after 1978 in Los Angeles was required to be exempt from rent control.) Finally, the Act prohibited local jurisdictions from imposing “vacancy control,” or requiring that below-market rents be maintained on rent-controlled apartments even after tenant turnover. In other words, Costa-Hawkins required that landlords of rent-controlled buildings be allowed to charge market-rate rent to new tenants moving into a unit that had been vacated (known as “vacancy de-control”). By repealing Costa-Hawkins, Prop. 10 would remove these three restrictions on the rent control policies that local jurisdictions are allowed to adopt.
What Is the Expected Impact of Proposition 10?
First, it is important to note what Prop. 10 would not do. Prop. 10 would not establish the basic authority of cities in California to choose to adopt rent control policies that apply to a substantial share of rental housing in the state. Local jurisdictions are already permitted to limit rent increases in all housing that is not explicitly exempted from rent control by Costa-Hawkins, and the older apartments that are not covered by Costa-Hawkins make up roughly half of all existing rental homes in California. In fact, most recently — in 2016 — voters in two San Francisco Bay Area cities approved new policies to establish local rent controls for the first time, to apply to rental apartments built before February 1995.[8] Moreover, if Prop. 10 passes, rent controls would not automatically be expanded to additional cities or (with a few exceptions) to additional rental homes in California.[9] In nearly all cases, cities or other local jurisdictions would have to take separate action, through their elected officials or voter initiatives, to adopt new local rent stabilization policies that would apply to the types of housing that are currently exempt from rent limits under Costa-Hawkins. Prop. 10 also would not force landlords to rent out units at a loss, since state law would continue to require that local rent control policies allow landlords to receive a “fair rate of return.”[10]
What Prop. 10 would do is allow cities to choose to limit rent increases within a broader range of rental homes, including single-family homes and apartments built since 1995 (or since earlier years in cities with longer-standing rent control policies) and/or when new tenants move into a vacant rental home. Cities that already have rent control policies in place (see Table 1) could choose to expand these policies to cover more of their rental stock or to restrict rent increases when a unit becomes vacant. Cities that do not currently have rent control policies could choose to adopt new policies that would apply to all of their rental housing or to any subset of rental homes without regard to the limitations currently set by Costa-Hawkins. For example, a city could choose to adopt new rent controls that would apply to all rental homes, or to all apartments built before 2005 but only for continuing tenants, or to both single-family homes and apartments built before 1995 but only for landlords that rent out at least 25 homes, or to any other subset of rental homes. Existing local rent control ordinances vary greatly in how they are designed, and any new local ordinances would likely also show considerable variation.
Table 1
While a large share of rental homes in California are not required to be exempt from rent control by Costa-Hawkins, a significant number of renters live in properties that could become newly covered by rent controls if Prop. 10 passes and their local jurisdiction chooses to adopt new rent control policies that apply to these homes. Statewide, more than one-third (35.5%) of renter households — or 2.06 million households — lived in single-family homes as of 2016, and this share has increased by about 10% since 2006, when 32.4% of renter households lived in single-family homes.[11] Renter households are most likely to live in single-family homes in the less urban, inland regions of the state — for example, more than half of renter households in the Sierra Nevada (55.4%), Central Valley (53.0%) and Far North (51.4%) lived in single-family homes in 2015-2016. Renters are least likely to live in single-family homes in the coastal urban regions of California — for example, in 2015-2016 less than a third of renter households occupied single-family homes in Los Angeles and the South Coast (29.3%) and the San Francisco Bay Area (30.6%).[12] These coastal urban regions are the areas where rents have been rising most rapidly, and while single-family homes represent a relatively smaller share of occupied rentals in these high-cost regions, they still house a significant number of renters in these areas.
Other renters live in apartments that were built within the last few decades. This means these units are not allowed to be covered by rent control under Costa-Hawkins, but could become covered if Prop. 10 passes and cities choose to adopt new rent control policies that include more recently built rentals. In the four largest cities that have current rent control policies in place — Los Angeles, San Francisco, San Jose, and Oakland — Costa-Hawkins limits rent controls to properties built before about 1980 (see Table 1). Across these four cities combined, 14.7% of renter-occupied apartments (198,000 homes) were built between 1980 and 1999, and another 8.5% (115,000 units) were built between 2000 and 2016 according to a Budget Center analysis of US Census data from 2016. If these cities chose to extend rent control to apartments built before 2000, for example, instead of before the year currently allowed by Costa-Hawkins, nearly 200,000 additional apartments could be covered by their rent stabilization policies.
How Could Low- and Moderate-Income Californians Be Helped by Proposition 10?
California’s housing affordability crisis most deeply affects low- and moderate-income households and renters.[13] Allowing local jurisdictions to expand rent control could provide more of California’s current renters with a guarantee of modest, predictable rent increases as long as they remain in the same home, protecting them from large or repeated jumps in rent that may outpace any increase in their incomes. This protection from rapidly rising rents is particularly valuable for lower-income households because wages and earnings for low- and midwage workers have experienced only sluggish growth in recent years, even as the economy overall has been improving.[14] Savings on rent due to rent control vary greatly depending on how local rent control ordinances are structured and how long a tenant has been in the same home, but they can be very large for long-term tenants in jurisdictions with strong control of rents. A study that examined renters in San Francisco from 1995 to 2012, for example, found that renters saved an average of $2,300 to $6,600 per person each year if they lived in a rent-controlled apartment.[15]
Rent control also encourages tenants to remain in the same home longer, both because tenants are less likely to be forced out by unaffordable rent increases and because they are less likely to be able to find equally low rent if they move to a new home. Increased housing stability is generally associated with positive health, social, and educational outcomes, particularly for children.[16] In some cases, rent control can discourage renters from moving to access new jobs or other opportunities or to secure a housing unit that better meets their needs — a less positive outcome — but this reduced mobility may be better understood as renters requiring a higher payoff from moving to compensate for the higher housing costs that would result from moving. At a neighborhood level, fewer moves among tenants translates into greater neighborhood stability, which can also mean a slower rate of gentrification in neighborhoods subject to gentrification pressures.[17][18]
Are There Potential Disadvantages for Low- and Moderate-Income Californians of Allowing an Expansion of Rent Control?
Local rent control policies can create incentives for landlords that have the undesired effect of working against the interests of low- and moderate-income renters. However, these potential negative effects can often be minimized with careful design of local rent controls and by coupling rent control with other local policies:
Capping allowed rent increases can incentivize landlords to neglect maintenance of their rental properties, but local rent control policies can address this issue by allowing landlords to pass through to renters some maintenance and improvement costs. Active local code enforcement can also help ensure that rent-controlled properties continue to meet health and safety standards.
Rent controls that apply to newly built housing, as Prop. 10 would allow cities to pursue, tend to reduce the expected profits from building market-rate rental housing, particularly in areas with rapidly rising rents, and could therefore discourage developers from building as much new rental housing as they otherwise would have. However, local rent control policies can address this problem by choosing to exclude newly constructed rental units from rent control for a period of years long enough to maintain an adequate incentive for developers. For example, newly built market-rate rental units could be exempted from rent control for 15 or 20 years after construction, allowing developers a set period of time in which to recoup their investment costs and collect potentially higher profits before the unit falls under rent control.
Extending rent control to new types of rental housing like single-family homes or more recently built apartments, as cities could choose to do if Prop. 10 passes, would tend to reduce the expected profits from renting out these homes, especially in areas with rapidly increasing rents. In turn, this would create an incentive for landlords with these types of properties to remove them from the rental market, converting them to ownership housing and selling them in order to cash out on the profits generated by a strong housing market. In fact, research shows that owners of rental properties subject to rent control are more likely to convert their properties to ownership housing.[19] Removing units from the rental market harms tenants in two ways: tenants residing in those units may be forced to move, and the reduction in the overall supply of rental housing will tend to intensify competition for vacant units and drive up rents in homes that are not subject to rent control. However, local jurisdictions can strive to minimize this outcome by coupling rent control with other tenant-protection policies. They can allow only “just cause” evictions and require significant compensation for tenants evicted when landlords seek to remove units from the rental market. They can also guarantee affordable legal services for tenants facing eviction to be sure these tenant protections are enforced. They can also restrict the conditions under which apartments can be converted to ownership condominiums, and implement separate policies designed to incentivize more housing development.
Of these potential negative effects of allowing an expansion of rent control, the removal of homes from the rental market may be the most likely to harm low- and moderate-income Californians. This is because a reduction in the overall number of rental units exacerbates the shortage of rental housing, which is a key driver of the rapid increase in market-rate rents, and also because local rent control and tenant-protection policies cannot fully prevent properties from being removed from the rental market. These negative effects would, however, be felt only by a subset of low-income renters: those who are seeking a new rental home or living in units not covered by rent control. They would therefore particularly affect individuals and families entering the local rental market for the first time (e.g., young adults moving out of their parents’ homes, or individuals moving to access job or educational opportunities, including those moving into California from out of state) and renters who are involuntarily forced to move to a new home (e.g., tenants whose homes are converted to ownership housing, or families in financial crisis who are evicted for nonpayment of rent). On the other hand, as noted above, expanding rent control would produce substantial benefits for continuing renters in homes newly subject to rent control who do not need or want to move. These substantial benefits for continuing renters must be weighed against the potential negative impacts for new renters and movers.
An expansion of local rent control policies would also likely reduce state and local government revenues, according to the Legislative Analyst’s Office (LAO), largely because the value of affected rental properties would be expected to drop, leading to lower property tax revenues over time.[20] Low- and moderate-income Californians benefit from the public services and supports funded by local and state revenues, so would be affected by a decline in these revenues. How much revenues decline would depend directly on how many local communities chose to extend rent control to newly allowed types of rental housing and how their rent control policies were designed. The drop in public revenues could range from insignificant to hundreds of millions of dollars annually, according to the LAO. This loss of revenues to support public services should be considered together with the substantial benefits to continuing renters from paying lower and more stable rents over time.
Conclusion: Understanding How Proposition 10 Relates to Other Policies to Address Housing Affordability
An expansion of local rent control policies, as Prop. 10 would allow, would not increase the supply of rental housing, and the shortage of rental housing is a root cause of California’s rapidly rising rents. If anything, as noted above, an expansion of rent control would be likely to encourage some landlords to remove current rental properties from the rental market, contributing to a decrease in the rental housing supply. For that reason, rent control expansion alone would not solve California’s housing affordability crisis. Other policies that are designed to increase the state’s supply of rental housing would also be needed, such as direct state and local investment in building affordable rental housing, financial or regulatory incentives to promote more private development of rental housing (such as inclusionary zoning or density bonuses), and local government accountability to support housing development at levels that meet the demand for rental housing. An expansion of rent control also would not address the needs of all renters, since rent control primarily benefits current renters who do not need or want to move to new homes, and may even disadvantage new renters and those who must or want to move.
However, many California renters are struggling to afford their housing costs now, and rent control is one of the few financially feasible and scalable policy tools available to address the immediate needs of those facing rents in the private market that are rising faster than incomes. Given the scale of the housing affordability crisis in California, policies that address affordability within housing provided by the private market are necessary, since the vast majority of renters will have to live in market-based housing. An expansion of rent control has advantages as a policy option to address private-market rents because it requires limited direct state and local costs and can be implemented and address housing affordability immediately. Local jurisdictions can also design local rent control policies in ways that minimize the potential negative effects of rent control and/or couple them with other policies that protect tenants and incentivize housing development.
Indeed, the design of local policies is central to the expected effects of Prop. 10 on housing affordability in California. If Prop. 10 passes, its effects would directly depend on how cities and other local jurisdictions chose to use their expanded authority to limit rent increases in rental housing at the local level. As a result, passage of Prop. 10 would primarily set the stage for local policymakers and voters to decide whether and how broadly to implement rent control policies within their local jurisdictions.
[8] The cities of Mountain View and Richmond passed rent stabilization measures in November 2016.
[9] The exception would be in cities with rent control policies still on the books from before the passage of Costa-Hawkins that apply to rental housing that is currently exempt from rent control under Costa-Hawkins. For example, the City of Berkeley’s pre-Costa-Hawkins rent control policy included vacancy control, and that portion of the ordinance was never repealed, so it would go back into effect immediately if Prop. 10 passes.
[10] Birkenfeld v. City of Berkeley, California Supreme Court, 17 Cal.3d 129 (1976).
[11] Unless otherwise noted, statistics in this section are from a Budget Center analysis of US Census Bureau, American Community Survey public-use microdata for California from 2015-2016, downloaded from IPUMS-USA (University of Minnesota, www.ipums.org).
[12] In terms of the remaining regions of California, renter households occupying single-family homes represented 48.7% of renters in the Inland Empire, 44.2% in the Sacramento Region, and 43.9% in the Central Coast in 2015-2016.
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
Proposition 2, which will appear on the November 6, 2018 statewide ballot, would allow California to move forward with a program, called No Place Like Home (NPLH), to finance the development of permanent supportive housing for Californians with mental illness who are homeless or at risk for chronic homelessness. The Legislature and Governor Brown placed Prop. 2 on the ballot because a lawsuit challenging NPLH and the state’s original financing plan has prevented California from implementing this new program. This post provides an overview of Prop. 2, discusses its expected impact, and examines other issues the measure raises in order to help voters reach an informed decision.
What Would Proposition 2 Do?
Prop. 2 asks California voters to approve a housing program — along with related financing — that state policymakers created in 2016, but which has been on hold due to litigation. No Place Like Home (NPLH) aims to develop permanent supportive housing for people with mental illness who are homeless or at risk for chronic homelessness. The state would finance this new housing using proceeds from the sale of up to $2 billion in bonds. These bonds would be repaid over several decades, with interest, using revenues from an existing state tax on California millionaires, which was imposed by Prop. 63 of 2004, the Mental Health Services Act. Currently, most of the proceeds from this tax — a 1% surcharge on taxpayers with annual taxable incomes of more than $1 million — pay for a broad range of mental health services that are provided or coordinated by California’s 58 counties. Voter approval of Prop. 2 would allow the state to use a portion of these Prop. 63 revenues — up to $140 million per year — to pay off the NPLH bond debt, likely over a 30-year period.
What Problem Does the No Place Like Home Program Aim to Address?
Safe and affordable housing is a key building block of health and well-being. Moreover, because housing is rooted in specific neighborhoods — each with its own unique mix of advantages and challenges — where people live helps to determine the opportunities that are available to them. These opportunities, in turn, influence each person’s educational, health, and economic outcomes. As one federal agency aptly puts it, without “a safe, affordable place to live, it is almost impossible to achieve good health or…one’s full potential.”
Unfortunately, California’s worsening housing crisis means that many people lack access to stable housing and find themselves living on the streets. As of January 2017, more than 134,000 residents of the Golden State — including both adults and children — were experiencing homelessness, according to the most recently published point-in-time count. This means that 34 out of every 10,000 Californians lacked a stable home, double the national rate of 17 per 10,000.
What’s more, a large share of people experiencing homelessness also struggle with mental illness. In January 2017, almost 34,700 homeless Californians — just over one-quarter of the total estimated homeless population — were identified as having a severe mental illness. Given the challenges of accurately gauging the true scope of homelessness, some research (here and here, for example) suggests that the actual share of homeless people with severe mental illness may be closer to one-third. Even more troubling: A sizeable share of those who are both homeless and mentally ill (half or more) grapple with drug or alcohol addiction stemming from their efforts to “self-medicate” in order to relieve their symptoms, according to experts.
No Place Like Home aims to assist Californians with mental illness who are homeless or at risk for chronic homelessness by building or rehabilitating permanent supportive housing specifically for this population. Supportive housing “is a highly effective strategy that combines affordable housing with intensive coordinated services to help people struggling with chronic physical and mental health issues maintain stable housing and receive appropriate health care,” according to one review of the literature. In other words, this approach provides affordable, long-term housing linked to wraparound services that can help people address mental health issues and other challenges.
Under NPLH, the state would borrow up to $2 billion and distribute nearly all of these funds to counties to both 1) finance the capital costs and 2) capitalize the operating reserves of permanent supportive housing.* In addition, counties would use other revenue sources — such as their annual Prop. 63 funds — to provide or coordinate services, including mental health and substance abuse treatment, for the tenants of supportive housing developments “for at least 20 years,” as required by the legislation that created NPLH in 2016 (Assembly Bills 1618 and 1628).
What Is the Expected Impact of the No Place Like Home Program?
No Place Like Home could significantly reduce the number of Californians with mental illness who are living on the streets. Assuming voter approval of Prop. 2, the state plans to award $262 million in NPLH funds each year for seven years, beginning at the end of 2018, the Legislative Analyst’s Office (LAO) has reported. This amount of funding could pay for the creation of “roughly 20,000 supportive housing units” over the course of a decade, with “a few thousand units” available by late 2020 or early 2021, the LAO estimates.
However, well over 30,000 Californians with severe mental illness are homeless, as noted above. This means that the projected 20,000 supportive housing units would fall short of the number needed to assist every Californian with severe mental illness who is experiencing homelessness. Therefore, even if voters approve Prop. 2, state and local leaders would still need to adopt additional policies targeting this population to fully address California’s overlapping crises of homelessness and mental illness.
Boosting the supply of permanent supportive housing could also decrease the use of other public systems by homeless residents with mental illness, in turn reducing state and local costs for these systems. For example, savings could come from lowering the number of homeless residents who end up in local jails or emergency rooms, according to a recent review of the research. Moreover, the evidence suggests that “the greatest reductions are likely achieved with supportive housing that focuses on people who are the costliest utilizers of services.”
Case in point: Los Angeles County’s “Housing for Health” (HFH) program. Launched in 2012, HFH uses permanent supportive housing to address the housing and health care needs of homeless residents, targeting “frequent users of health care services,” according to a RAND evaluation. RAND’s key finding: HFH “reduced health care use and county costs.” Even after taking into account the cost of supportive housing, LA County saved $1.20 (from reduced health care and other social service costs) for every $1 invested in the program.
What Are the Tradeoffs in Using Bond Dollars to Build Supportive Housing Through the No Place Like Home Program?
Under No Place Like Home, the state would issue up to $2 billion in bonds, with the proceeds going to build permanent supportive housing for homeless Californians with mental illness. These bonds would be repaid, with interest, from annual revenues that are generated by Prop. 63’s “millionaire’s tax,” which provides funding for mental health services that are delivered or coordinated by counties. As outlined in AB 1628 of 2016, the state would use up to $140 million per year in Prop. 63 revenues to pay the debt service (principal + interest) on the bonds. The most recent state estimate assumes a 30-year debt-service schedule with a 4.2% interest rate, resulting in projected payments of approximately $120 million per year.
By comparison, Prop. 63’s tax on millionaires typically generates over $1 billion per year for mental health services, and the state expects to collect more than $2 billion from the tax in 2018-19, the fiscal year that began on June 30. (Proceeds from this tax fluctuate significantly from year to year.) As a result, a relatively small share of Prop. 63 revenues would be used to pay debt service on the NPLH bonds in any given year. For example, if Prop. 63 generated $1 billion in revenues, the state’s projected payment would equal 12% of these funds ($120 million / $1 billion). If, in another year, Prop. 63 raised $2 billion in revenues, the state’s projected payment would amount to 6% of these funds ($120 million / $2 billion).
Issuing bonds would allow the state to quickly amass a large amount of funding (up to $2 billion) to jump-start the development of permanent supportive housing around the state. Boosting the supply of supportive housing over a relatively short period would allow counties to better focus their resources on an otherwise hard-to-serve population — those with mental illness who are homeless or at risk for chronic homelessness — while potentially achieving improved outcomes for this population through the provision of long-term housing combined with mental health and other supportive services.
However, in selling bonds, California would incur a debt that would have to be repaid with interest. As a result, over time the cost of servicing the debt would far exceed the amount of the borrowed funds. Using the example cited above, debt service on the NPLH bonds could amount to $120 million per year, assuming a 30-year debt-service schedule with a 4.2% interest rate. Over three decades, these payments would total $3.6 billion — much higher than the original $2 billion bond issuance. If the state used the maximum allowable amount of Prop. 63 revenues for NPLH ($140 million per year), then the debt service would total $4.2 billion over 30 years — more than double the amount of the original bond issuance. Moreover, because annual NPLH bond payments would be funded with Prop. 63 revenues — which primarily go to county mental health programs — counties would receive well over $100 million less in Prop. 63 funding each year to meet the mental health needs of their residents.
Yet, Californians who are struggling with mental illness and living on the streets are among the most vulnerable people in the state — a reality that is explicitly recognized in Prop. 63 (see Section 2(d) of the initiative’s “Findings and Declarations”). What’s more, counties have long been responsible for assisting Californians with severe mental illness and collectively receive billions of dollars each year to do so. While counties’ annual mental health funding would be reduced if voters approve Prop. 2, counties could also experience savings in other public systems, such as jails, due to the expansion of supportive housing, as explained above. Counties could use such savings to expand or enhance other local services, including services for residents of supportive housing. Moreover, counties could bolster their support for mental health services by prudently allocating the large amounts of unspent Prop. 63 funds that have been allowed to accumulate at the local level due to what the State Auditor recently called the state’s “ineffective oversight of local mental health agencies.”
What Do Proponents Argue?
Proponents of Prop. 2 include the California Police Chiefs Association, the California State Association of Counties, the League of California Cities, and Mental Health America of California. Proponents argue that Prop. 2 “delivers the proven solution to help the most vulnerable people experiencing homelessness in California” by “build[ing] housing and keep[ing] mental health services in reach for people — the key to alleviating homelessness complicated by mental illness.”
What Do Opponents Argue?
The National Alliance on Mental Illness (NAMI) Contra Costa opposes Prop. 2. NAMI Contra Costa argues that Prop. 2 “takes Billions [of dollars] away from our loved ones and rewards developers, bond-holders, and bureaucrats.” This organization further states that “counties already know where to best acquire housing for access to critical services. Prop. 2 cuts off local input and predetermines the balance between treatment and housing needs.”
Conclusion
Our state’s worsening housing crisis has left more than 130,000 Californians homeless, including tens of thousands of people with severe mental illness. Prop. 2 would take a significant step toward reducing the number of people with mental illness who are living on the streets. It would do so by allowing the state to sell up to $2 billion in bonds and use the proceeds to spur the creation of permanent supportive housing (stable housing linked to services) for this population. These bonds would be repaid over several decades with revenues from an existing state tax on California millionaires that was imposed by Prop. 63 of 2004. Prop. 63 directs most of the funds raised by this “millionaire’s tax” to county mental health programs. If voters approve Prop. 2, up to $140 million per year in Prop. 63 revenues would be used to pay off the new bond debt, leaving less for mental health services.
A key question for voters is whether the benefits of using bond funds to develop up to 20,000 supportive housing units around the state outweigh the potential impact of a relatively small decrease in counties’ annual mental health funding. On the one hand, increasing the supply of supportive housing for homeless residents with mental illness would allow counties to better focus their resources on an otherwise hard-to-serve population while potentially improving outcomes for these individuals. On the other hand, counties would receive less annual Prop. 63 funding to address the mental health needs of their residents.
Two key factors would mitigate the impact of this Prop. 63 funding reduction on local mental health services. First, developing supportive housing for homeless residents with mental illness could decrease the use of other public systems, such as jails, thus reducing state and local costs for these systems. At least some of the resulting savings would accrue to counties, which could shift these freed-up revenues to other services, including mental health treatment. Second, many counties have amassed exceedingly large sums of unspent Prop. 63 funds that could be drawn down over a number of years in order to bolster annual support for mental health services.
* Up to 5% of the total bond funds could be used for state administrative expenses. In addition, up to 4% of the bond funds that would be allocated to counties through NPLH’s “competitive program” could be used to create default reserves.
There was a problem processing your signup. Please try again. Or contact us
Please check your email to confirm your signup.
This website uses cookies to analyze site traffic and to allow users to complete forms on the site. The California Budget & Policy Center does not share, trade, sell, or otherwise disclose personal information. By using our website you agree to our Privacy Policy.