CITY AND COUNTY OF SAN FRANCISCO BOARD OF SUPERVISORS BUDGET AND LEGISLATIVE ANALYST 1390 Market Street, Suite 1150, San Francisco, CA 94102 (415) 552-9292 FAX (415) 252-0461 Policy Analysis Report To: From: Re: Supervisor Preston Budget and Legislative Analyst’s Office Estimate of unpaid residential rent in San Francisco due to COVID-19 pandemic and related public health orders October 27, 2020 Date: SUMMARY OF REQUESTED ACTION Your office requested that the Budget and Legislative Analyst estimate the amount of unpaid residential rent in the City and County of San Francisco due to the COVID-19 pandemic, i.e. from direct health impacts and indirect economic impacts related to job or income loss. For further information about this report, contact Fred Brousseau at the Budget and Legislative Analyst’s Office. Executive Summary  The onset of the COVID-19 pandemic and the March 2020 sheltering-in-place orders in San Francisco and California had an immediate impact on the economy. Unemployment in San Francisco increased from a rate of 2.3 percent in February 2020 to 12.6 percent in April 2020. This translated to an increase from 13,299 individuals unemployed in February (and similar numbers in the preceding six months) to 69,806 in April, or an increase of 56,507 individuals.  One impact of the rapid spike in unemployment in March and April was an increase in partial or full non-payment of residential rent by San Francisco tenants. While unemployment decreased somewhat in July and August it remained well above pre-pandemic levels. The reduction in residential rent paid in San Francisco has been documented by various organizations including the Census Bureau, the San Francisco Apartment Association, and the Terner Center for Housing Innovation at UC Berkeley. Along with the Census Bureau, the National MultiFamily Housing Council and the National Association of Hispanic Real Estate Professionals have reported the same phenomenon at the national level.  To estimate the amount of unpaid rent in San Francisco since sheltering-in-place began in March 2020, we reviewed datasets, survey results, and other analyses from various sources including those identified above. While no one source provides a definitive answer to the question of the exact amount of residential Budget and Legislative Analyst 1 Memo to Supervisor Preston October 27, 2020 rent that has gone unpaid in San Francisco since March, we found similarities in the results of the various sources of information that allowed us to prepare a range of compatible estimates using varying assumptions about the proportion of newly unemployed that reside in and pay for rental housing.  We prepared a range of estimated unpaid residential rent amounts in San Francisco based on these various sources, ranging from $13.6 million a month up to $32.7 million per month for the period April - September 2020, as shown in Exhibit A.  The primary difference between our low and high estimates is that the low estimate assumes that those unemployed due to COVID-19 are distributed by housing type proportional to the City’s overall population, with 13,765 renter households affected, or approximately six percent of all renter households; our high end estimate assumes that a disproportionate number of renter households, 33,200, are affected by COVID-19 related unemployment, or 15 percent of all renter households. The estimated unpaid rent is shown for both scenarios for one, three, and six months in Exhibit A. Exhibit A: Low and High Estimate of Residential Rent Partially or Not Paid by San Francisco Renters, April – September 2020   Estimate # of Renter Households 1 month 3 months 6 months Low 13,765 $13,558,128 $40,674,384 $81,348,768 High 33,200 $32,702,000 $98,106,000 $196,212,000 Months of Unpaid Residential Rent  For both our low and high unpaid rent estimates shown in Exhibit A, we assume median monthly rent of $1,970 and that unpaid rent for these households overall totaled 50 percent of the total amount due. This approach accounts for some households paying no rent, some paying partial rent, and some paying the full amount with other resources available to them such as savings, loans, basic and expanded unemployment insurance, and CARES Act stimulus payments.  Other indicators of tenants having difficulty paying rent were captured by input from a number of San Francisco community-based organizations that provide support services such as rental assistance and counseling for tenants facing eviction or having other issues with their rentals. The following four organizations surveyed all reported increases starting in March 2020 in the number of clients or Budget and Legislative Analyst 2 Memo to Supervisor Preston October 27, 2020 applicants seeking residential rental assistance or support services due to their inability to partially or fully pay their rent: Eviction Defense Collaborative, Housing Rights Committee, Causa Justa::Just Cause, and the Mayor’s Office of Housing and Community Development’s Give2SF Housing Stabilization Program.  Our low and high estimates of unpaid residential rent between April and September 2020 may not be good predictors of future unpaid rent. While unemployment in the City declined in August, though still remained well above February and prior months, protections and supports during the early months of the pandemic that assisted some tenants in paying their rent may be diminished or exhausted completely in the coming months. These supports include personal savings, loans, regular and expanded unemployment insurance, and stimulus payments made available through the CARES Act.  In addition to unpaid rent, landlords face a potential loss of income related to the pandemic. Renter property owners report increases in notices from tenants who wish to vacate their units, break their leases, or are requesting rent reductions. San Francisco Apartment Association surveys of its members since April 2020 have found landlords reporting increases in such notices as well as tenants breaking their leases or unexpectedly giving notice to vacate.  Financial assistance has been available to tenants in a number of forms since the onset of the pandemic and sheltering-in-place: regular unemployment insurance, and expanded unemployment insurance benefits and stimulus payments funded by the CARES Act, the latter two of which have largely expired. At the local level, the primary support service provided by the City has been the Give2SF Housing Stabilization Program, most of the funding for which has been expended as of the writing of this report. Unless new programs are adopted at the federal or state level, unpaid rent is likely to increase as long as San Francisco renters remain unemployed due to COVID-19.  Anti-eviction protections remain in place for several months but unpaid rent will become due in March 2021 under current legislation in many cases and rent is due on time starting again in February 2021. For tenants who have been and will remain unemployed in the coming months, there is little rental relief available. Project staff: Cody Xuereb, Fred Brousseau Budget and Legislative Analyst 3 Memo to Supervisor Preston October 27, 2020 Table of Contents 1. The Impact of COVID-19 on San Francisco’s Rental Market ................... 4 2. Estimates of Total Unpaid Rent due to COVID-19 ................................ 12 3. Summary of Legal Protections and Financial Assistance Available for Tenants and Landlords .......................................................................... 20 Appendices A. Detailed Tables with Non-Payment of Rent Estimates B. Detailed Explanation of Methodology for Calculating Unpaid Residential Rent Estimates C. Detailed Breakdown of Unpaid Estimates by Building Size and Scenario 1. The Impact of COVID-19 on San Francisco’s Rental Market The COVID-19 pandemic has had significant direct and indirect impacts on the economic, physical and mental wellbeing of all San Francisco residents. Various government, academic, and private sector data and reports show that the pandemic has resulted in increased unemployment, reductions in income, and non-payment of rent nationally, statewide, and in San Francisco. A number of studies we reviewed show that renters have been particularly impacted due to the significant economic impacts the pandemic and related public health orders have had on employment and earnings. Given the existing vulnerability of renters due to the City’s housing affordability crisis, the COVID-19 pandemic has added additional pressure due to its disproportionate impact on low-income working households. These studies also highlight disparate impacts on people of color, including Latinx and African American populations, related to higher COVID-19 exposure rates and employment in impacted industries.1 COVID-19 Confirmed Cases and Unemployment in San Francisco 1 An April 2020 UCSF study of COVID-19 infections in San Francisco’s Mission neighborhood found the infection rate was 11 times higher than the city average and 95 percent of those who tested positive were Latinx, despite making up 44 percent of those tested. Nationally, Black and Latinx people were 3.8 and 2.5 times more likely to die from COVID-19 than White people, respectively (APM Research Lab, July 2020). People of color are also more likely to work in high-contact service industries and have pre-existing health conditions. Budget and Legislative Analyst 4 Memo to Supervisor Preston October 27, 2020 Exhibit 1 below provides an overview of the number of new COVID-19 cases in San Francisco alongside the number of unemployed individuals. This shows the significant increase in unemployment following the declaration of the first Bay Area Stay at Home order on March 16 2020, with the number of unemployed rising from 13,299 in February and 18,065 in March to 69,806 in April. While this has improved somewhat through August following gradual reopening, the latest county level unemployment data indicates 49,845 individuals were out of work in August, an increase of 31,790 individuals compared to March. Exhibit 1: COVID-19 Confirmed Cases and Unemployment in San Francisco, August 2019 – September 2020 Unemployed (#) SF COVID Cases 3,500 3/16: SF Stay at Home Order Issued 70k 70k 70k 71k 3,000 63k 60k 2,500 50k 50k 2,000 40k SF COVID Cases (new) No. Unemployed (thousands) 80k 1,500 30k 20k 1,000 18k 14k 11k 12k 12k 11k 13k 13k 500 10k 0k 0 Aug Sep Oct Nov Dec Jan Feb Mar 2019 Apr May Jun Jul Aug Sep 2020 Source: SF Department of Public Health, US Bureau of Labor Statistics Local Area Unemployment Statistics Notes: Unemployment data for San Francisco County is published 6-8 weeks following the end of the month. This unemployment data may be an underestimate of those impacted by job or income loss as it does not include those with reduced hours or on furlough. For example, according to the California Employment Development Department, the total number of individuals claiming Unemployment insurance benefits in San Francisco County was 78,287 in July, indicating potential undercounting due to individuals leaving the Budget and Legislative Analyst 5 Memo to Supervisor Preston October 27, 2020 workforce, not actively looking for work, or working part time.2 The latest data from the U.S. Census Bureau’s Household Pulse Survey3 also indicates that 52 percent of people over 18 in the San Francisco Metropolitan Area reported a loss of income in their household since the start of COVID-19, as of the end of September.4 San Francisco Rental Market during the COVID-19 Pandemic Increase in Renter households not Fully Paying Rent While there is no single source of data on the extent of non- or underpayment of residential rent over time, several governmental and nongovernmental organizations have developed surveys and carried out analyses to measure this issue. Some of the organizations and efforts discussed further below include:  US Census Household Pulse Survey: representative weekly and biweekly national survey to measure household experiences during the COVID-19 pandemic, including rent payment.  University of California, Berkeley Terner Center for Housing Innovation: analysis of Household Pulse Survey and other surveys to estimate the impact of job loss on payment of rent.  San Francisco Apartment Association: monthly survey of Association members on payment of rent and other issues.  U.S. Bureau of Labor Statistics (BLS): unemployment statistics for San Francisco allow for estimates of the number of unemployed individuals since the onset of the pandemic and imputation of the number of associated households paying partial or no rent. Exhibit 2 below provides a summary of the findings on non-payment of rent for each of these organizations, as well as a summary of our imputation of non-payment using BLS unemployment statistics for San Francisco. While the methodology, robustness and coverage of these sources differ, the estimates covering San Francisco and the region range from approximately 2 California Employment Development Department, Unemployment Insurance Claimants by County (https://edd.ca.gov/about_edd/Quick_Statistics_Information_by_County.htm) 3 Since May 2020, the U.S. Census Bureau has run a representative weekly and biweekly national survey to measure household experiences during the COVID-19 pandemic, including information on employment and housing. The survey design allows representative estimates produced at the national and state level as well as for the top 15 largest Metropolitan Statistical Areas. 4 U.S. Census Household Pulse Survey, Week 15 (Sept. 16 – 28). San Francisco Metropolitan Statistical Area includes San Francisco, San Mateo, Alameda, Contra Costa, and Marin counties. Budget and Legislative Analyst 6 Memo to Supervisor Preston October 27, 2020 4.2 to 15 percent of renter households not fully paying their rent. This range likely reflects differences in the survey design, question phrasing, and individuals surveyed. Given the robustness and lower-level data available for the BLS unemployment statistics and Terner Center, we focused our analysis on these two sources and use estimates based on these sources for our range of two estimates of the amount of unpaid rent. Exhibit 2: Summary of Estimates of Percentage of Renter Households’ Non-Payment of Rent, by Data Source, April – September 2020 Source Survey Measure Coverage Unit Apr May Jun Jul Aug Sept BLA Imputation using BLS data % of Renter Households paying partial or no rent (based on # of unemployed individuals) SF County Renter Households 6.5% 6.6% 6.7% 5.8% 4.2% - Terner % of Renter Households Impacted by Job Loss SF County Renter Households 15.0%+ 15.0%+ 15.0%+ - - - HPS % Renter Households not Paying Last Month's Rent on Time SF MSA** All People 18+ in Rental Units - 13.7% 11.7% 11.1% 12.4% 11.9% SFAA % Renter Households not Paying any rent due to COVID-19 SF County Rental Units 5.7% 3.2% 2.3% 3.1% 2.2% - SFAA % Renter Households Paying Partial Rent SF County Rental Units - - 2.5% 3.4% 4.0% - SFAA Total % Renter Households not Paying or Partially Paying Rent SF County Rental Units 5.7% 3.2% 4.8% 6.5% 6.2% - Sources: BLS = BLA calculations using unemployment data from U.S. Bureau of Labor Statistics Local Area Unemployment Statistics; HPS = US Census Bureau’s Household Pulse Survey; Terner = UC Berkeley Terner Center for Housing Innovation, Analysis of BLS Current Employment Statistics and American Community Survey; SFAA = San Francisco Apartment Association, Survey of Association Members. Notes: See supporting tables in Appendix A. *Net Year over Year difference, 2019 vs. 2020. ** San Francisco Metropolitan Statistical Area includes San Francisco, San Mateo, Alameda, Contra Cost and Marin counties. + 15% estimate is a cumulative total of renter households impacted by job loss from April to June 2020. A review of the sources above and additional research carried out at the national level indicates that the key factors and characteristics associated with non-payment of rent include:  Job or income loss: A representative survey of 1,000 renters in Los Angeles County by UCLA and USC found that those who lost their job or Budget and Legislative Analyst 7 Memo to Supervisor Preston October 27, 2020 income were 2.5 to 4.5 times more likely to have trouble paying rent.5 The Census Bureau’s Household Pulse Survey results indicate that Californians who experienced income loss were 2.7 times more likely to report having no confidence in making the next month’s rent (as of September 16 – 28).  Lower income households (i.e., those earning under $50,000 per year): an analysis by the Terner Center for Housing Innovation at UC Berkeley estimated that at the time of their job loss, 47 percent of renter households earned 80 percent of Area Median Income or less in San Francisco.6 The UCLA/USC survey found non-payment of rent was twice as likely among those earning under $25,000 compared to all renters. 7  Race/ Ethnicity: Black and Hispanic renters were found to be almost twice as likely to make a partial rent payment from May to July 2020 compared to White renters in the UCLA/ USC Los Angeles renter survey. The Terner Center estimated that two thirds of renters who experienced job loss in San Francisco from April to June had a Person of Color in the household.  Building Type: Households in single family residences and smaller buildings (2 – 4 units) appear to be twice as likely to be behind on rent relative to those in mid-size buildings with 5 – 49 units, according to a Terner Center Analysis of the Census Bureau’s Household Pulse Survey. However, renters in buildings with 50 or more units were also more likely to be behind on rent payments than those in mid-size buildings.8 Two sources of information reviewed for this report documented partial or non-payment of rent specifically in San Francisco since sheltering-in-place began in March 2020. First, the national weekly U.S. Census Bureau Household Pulse Survey asks respondents, including San Francisco residents, if they paid last month’s rent timely. The second source is the San Francisco Apartment Association which has been surveying its members monthly since April 2020 regarding the number of tenants paying partial or no rent as a 5 Manville, M., et al., COVID-19 and Renter Distress: Evidence from Los Angeles (UCLA-USC Report published August 31 2020). 6 Terner Center for Housing Innovation, “COVID-19 and California’s Vulnerable Renters” (Blog post dated August 4, 2020). 7 Manville, M., et al., COVID-19 and Renter Distress: Evidence from Los Angeles (UCLA-USC Report published August 31 2020). 8 Terner Center for Housing Innovation, “The ongoing housing crisis: California renters still struggle to pay rent even as counties re-open” (Blog post dated October 2, 2020). Budget and Legislative Analyst 8 Memo to Supervisor Preston October 27, 2020 result of COVID-19. The Association also surveys its members monthly about other metrics such as notices to vacate filed by tenants, discussed further below. It is also worth noting that the estimates of non-payment are also likely to underestimate the number of undocumented residents who are less likely to be captured in official surveys. These individuals are likely at higher risk of non-payment due to limited access to federal and state benefits. A 2016 analysis by the Migration Policy Institute estimated there could be up to 49,000 unauthorized immigrants living in San Francisco County.9 US Census Bureau’s Household Pulse Survey Starting in May 2020, the U.S. Census Bureau developed a national survey to measure household experiences during the COVID-19 pandemic. The survey includes questions on employment, the ability to meet basic needs, such as food and rent, and income sources. The Survey includes a question about whether individuals in renter households are current on their last month’s rent payment for the San Francisco and Los Angeles Metropolitan Statistical Areas as well as for California and the U.S. overall. While San Francisco renters appear to be faring better than these other areas, an average of 12 percent of individuals in renter households reported not being current on rent payments between May 5th and September 28th, 2020. This is lower than the averages for California (15 percent), nationally (16 percent), and the Los Angeles metro area (17 percent). San Francisco Apartment Association Surveys This San Francisco Apartment Association has been conducting monthly surveys of its members since April 2020 requesting data on: unpaid rent, partially paid rent, notices to vacate filed, requests for rent reductions, and other metrics. The organization’s survey results from April through August 2020 show that between 3.2 and 6.5 percent of renter households in San Francisco paid partial or no rent due to COVID-19. 9 Migration Policy Institute (2016), Profile of the Unauthorized Population: San Francisco County, CA Budget and Legislative Analyst 9 Memo to Supervisor Preston October 27, 2020 Other Local Indicators of Non-Payment of Rent: Community Based Organizations and MOHCD Rent Relief Program Community Based Organization Feedback We contacted several community-based organizations (CBOs) that help tenants facing evictions about the demand for their assistance related to non-payment of rent since sheltering-in-place began.10 We received the following information from the CBOs we contacted, confirming the trend reported above that tenants who have lost their jobs or been cut back are facing the prospect of incurring debt by not paying or partially paying their rent.  Eviction Defense Collaborative: the EDC reported a doubling in requests for counseling and rental assistance since the start of the pandemic. They have also disbursed $584,000 in rental assistance to 191 individuals as of September 23 (most of this through the Give2SF Housing Stabilization Program).11  Housing Rights Committee of San Francisco: since the Stay at Home order in mid-March through the end of September, HRCSF estimates they received 1,230 inquiries from tenants related to inability to pay rent, around 75 percent of all inquiries received. HRCSF reports that this is a significant increase from before the start of the pandemic when only a few calls per month were received regarding difficulty paying rent.  Causa Justa::Just Cause: since the end of March through the end of September, Causa Justa reported receiving 157 calls from tenants unable to pay rent due to COVID-19, representing around one in five of all calls received over this time. San Francisco’s Give2SF Housing Stabilization Program In May, the Mayor’s Office of Housing and Community Development (MOHCD) launched a Housing Stabilization Program to provide rental assistance to tenants facing eviction due to financial hardship caused by the COVID-19 pandemic. The program aims to prevent evictions by providing financial assistance for future rents not covered by emergency eviction 10 The full list of CBOs contacted includes: Chinatown Community Development Center, Causa Justa::Just Cause, Housing Rights Committee of San Francisco, San Francisco Tenants Union and the Eviction Defense Collaborative. 11 The Eviction Defense Collaborative is one of five organizations providing rental assistance through the Give2SF Housing Stabilization Program. Budget and Legislative Analyst 10 Memo to Supervisor Preston October 27, 2020 protections, rather than past due rent.12 Financial assistance is provided mostly to third parties (such as landlords) and accompanied with a written communication to landlords and tenants to ensure funds are applied for the month(s) specified. From May 4 to September 30, the program received 9,200 applications requesting a total of $43 million in housing assistance. This equates to roughly $4,700 per applicant on average, or around 2.5 months’ rent13. In a July 2020 proposal to the City’s Economic Recovery Taskforce, MOHCD estimated that up to 7,650 renter households at high risk of homelessness may need financial assistance for COVID-19-related hardship at an estimated total cost of $50 million. The program is primarily funded by private donations to the City’s Give2SF fund as well as federal funding from the CARES Act.14 According to MOHCD, the program currently has $7.5 million in direct financial assistance funding available, of which $6.5 million has been committed to help 1,449 households (around $4,500 per household on average). Around half of the committed total has been paid. Ongoing funding from private donors for this program at the current level is not guaranteed. Affordable Housing Providers As part of our review, we also spoke with MOHCD regarding non-payment of rent experienced by affordable housing operators regulated by the MOHCD. MOHCD staff indicated data collection to assess the impact of non-payment on the financial condition of housing operators was ongoing and expected to conclude by February 28, 2021. MOHCD is also working with affordable housing providers to amend older ground leases and loans for 40 properties to allow providers greater access to surplus residual receipts, in line with MOHCD’s current residual receipts policy. A national survey of 164 major rental operators (including 70 percent with mostly subsidized units) in August 2020 found that 89 percent had experienced a reduction in revenues related to the pandemic, with an 12 For example, if an emergency tenant protection makes September missed rent un-evictable but not October missed rent, then the program would provide financial assistance to cover October, rather than the arrearage for which there is eviction protection. 13 Based on a median gross rent of $1,970 for San Francisco County based on Terner Center analysis of the 2018 American Community Survey. 14 Community Development Block Grant COVID-19 funding. Budget and Legislative Analyst 11 Memo to Supervisor Preston October 27, 2020 average 11.8 percent decline in revenues.15 Among those experiencing revenue declines, 73 percent of providers reported non-payment of rent as the driver of revenue declines. While most affordable housing providers regulated by MOHCD are required to maintain a 25 percent operating expense reserve, an extended period of high unemployment combined with limited additional federal assistance for renters could lead to providers delaying maintenance or reducing other operating services to make up for revenue shortfalls. Housing providers have also seen an increase in expenses to meet public health orders (i.e. Personal Protection Equipment, reconfiguration of physical spaces to meet social distancing requirements, etc.). 2. Estimates of Total Unpaid Rent due to COVID-19 Despite multiple estimates of the number of renter households that are behind on rent payments or experiencing job loss due to COVID-19, we were unable to find any estimates of the amount of unpaid rent at the local or regional level, including for the City and County of San Francisco. Given these limitations, we used several different data sources to produce a range of estimates of the amount of unpaid rent among San Francisco renters between the onset of sheltering-in-place through September 2020. Our resulting estimates of unpaid rent between April and September 2020 range from $13.6 million up to $32.7 million per month. For the six month period, this equates to between $81.3 million and $196.2 million. Our range of estimates reflects different assumptions including, on the low end: (1) that the pandemic-related unemployed are distributed between renters and homeowners proportional to their distribution in the population of San Francisco overall (i.e., renters do not make up a disproportionate share of the unemployed), (2) that these unemployed individuals are members of households with 2.16 occupants each, or the average for San Francisco renter households, and (3) these renter households with unemployed individuals paid 50 percent of their rent on average. The latter assumption is to account for partial or full rent paid by some households with personal savings, extended unemployment insurance and stimulus 15 Affordable Housing Finance, “Affordable Housing Owners See Higher Operating Costs Due to COVID-19” (article posted October 2, 2020). Budget and Legislative Analyst 12 Memo to Supervisor Preston October 27, 2020 payments from the CARES Act, loans, funds from rental assistance programs, and other sources.16 Our high end estimate assumes that renters have disproportionately suffered unemployment and reduced employment based on the distribution of industries and occupations affected by the pandemic. This results in a disproportionate share of renter households assumed to have not paid or only partially paid their rent. We have adjusted the resulting maximum unpaid rent for those households to 50 percent of the total amount due for the same reasons as described above for our low-end estimate. Exhibit 3: Low and High Estimate of Rent Partially or Not Paid by San Francisco Renters, April – September 2020  Estimate Months of Unpaid Residential Rent # of Renter Households 1 month 3 months 6 months Low 13,765 $13,558,128 $40,674,384 $81,348,768 High 33,200 $32,702,000 $98,106,000 $196,212,000 Sources: BLA calculations based on BLS Local Area Unemployment Statistics and UC Berkeley Terner Center analysis. Notes: Assumes median monthly rent of $1,970 based on Terner Center analysis of 2018 ACS data. We do not assume that the estimated amounts of unpaid rent shown in Exhibit 3 will continue to be unpaid in coming months as, at this time, the safeguards in place during the first six months of the pandemic including enhanced unemployment insurance and stimulus payments through the CARES Act have now expired and, as of the writing of this report, no replacements of the same magnitude are in place. Further, to the extent that renter households have used personal savings and/or loans or funds from rental assistance programs, these amounts could be depleted and will not necessarily be replenished. This indicates that the amount of unpaid rent could likely increase in coming months if the pandemic and sheltering-inplace remains in effect and unemployment remains well above average. A decrease in unemployment in August 2020 could have the effect of reducing the amount of unpaid rent, though the number of unemployed in August in San Francisco still greatly exceeded pre-pandemic levels. 16 This 50 percent assumption is also similar to estimates used in a report on unpaid rent due to COVID-19 nationally for the National Council of Social Housing Agencies by STOUT from September 28, 2020. Budget and Legislative Analyst 13 Memo to Supervisor Preston October 27, 2020 Low estimate of partially or unpaid rent: $13.6 million per month Our low end estimate of partially or unpaid rent between April and August 202017 is based on reported unemployment in San Francisco during that period. As shown in Exhibit 4, the incremental number of unemployed individuals in San Francisco after sheltering-in-place began amounted to an average of 51,646 for that period. We assumed that 58 percent of those individuals, or 29,732 individuals, were renters, consistent with the general distribution of individuals among those living in renter and owner-occupied housing units in San Francisco as a whole18. We then assumed that those individuals represented 13,765 households, based on the average renter household size in San Francisco of 2.16.19 Finally, we assumed that, on average, these renters paid 50 percent of the maximum amount of rent due to account for partial, no, and full payments by some renters even though they were unemployed. Exhibit 4: Assumptions for Low End Estimate of Amount of Unpaid Rent in San Francisco, April – August 2020 Assumptions/ Inputs Average Number of San Franciscans Unemployed April – August 2020 Average Incremental Unemployed after Sheltering-in-Place Number of Renters (@ 58% of all individuals) Total Number of Renter Households experiencing Job Loss (@ average 2.16 persons/household) Share of Total Renter Households in San Francisco Value 64,945 51,646 29,732 13,765 6.0% Median Rent for Impacted Renter Households $1,970 Estimated Unpaid Rent (50% of maximum due) $13,557,540 Sources: BLA analysis of BLS Local Area Unemployment Statistics and American Community Survey (ACS) 2019 1-year estimates (for renter and household size assumptions). As shown in Exhibit 4, our low estimate assumes that six percent of renter households in San Francisco did not fully or partially pay their rent between April and August 2020. As shown below, our high estimate is that 33,200 renter households, or 15 percent of all renter households, did not pay rent during the first months of sheltering-in-place. Our high estimate reflects the assumption that renters have been disproportionately affected by unemployment during the pandemic. 17 September unemployment data was not available at the time of preparation of this report. American Community Survey 2019, Table B25008 19 American Community Survey 2019, Table B25010 18 Budget and Legislative Analyst 14 Memo to Supervisor Preston October 27, 2020 Other data sources and analysis that we reviewed present a range of renter households not paying some or all of their rent during sheltering-in-place months to date including an average of 5.8 percent of renter households based on San Francisco Apartment Association surveys of their members up to an average of 12.1 percent from the Census Bureau’s Household Pulse Survey conducted after sheltering-in-place began.20 Our low and high estimates incorporate the range of these estimated shares of renter households not paying some or all of their rent. Limitations of our low estimate are that it could be understating the actual amount of unpaid or partially paid rent between April and August 2020. The unemployed due to COVID-19 could be disproportionately renters, for which this estimate is not adjusted, and there would thus be a higher number of households not paying or only partially paying their rent. Additionally, unemployment statistics focus on those who lost their job and does not include those who had a reduction in work hours or income due to the pandemic. Conversely, the estimates also do not account for replacement income received through unemployment insurance benefits. High estimate of partially or unpaid rent: $32.7 million per month For our high end estimate of unpaid rent, we combined an analysis by the Terner Center of renter households that experienced job loss from April to June 2020 with data from other sources on the extent of non-payment of rent across time. These estimates should be considered approximate given limitations in the data used and assumptions made (discussed below). We have also produced comparisons for these estimates by using other data sources and methodologies to estimate rent arrears at the state and national level (see Appendix B). We believe this approach may overestimate the total extent of unpaid rent but is likely the most robust given the documented link between loss of employment and ability to pay rent discussed above. The Terner Center Analysis was produced by combining monthly unemployment data from the Bureau of Labor Statistics’ Current Employment Survey (CES) from April to June 2020 with more detailed survey data from the 2018 U.S. Census Bureau’s American Community Survey (ACS). As a first step, unemployment by industry in the CES was used to estimate rates of unemployment by industry among renters identified in the ACS. Based on these industry estimates, unemployment could then be randomly 20 Applied to renter household population in the San Francisco Metropolitan Statistical Area (San Francisco, San Mateo, Alameda, Contra Cost and Marin counties. Budget and Legislative Analyst 15 Memo to Supervisor Preston October 27, 2020 assigned among renters in the ACS using a weighting factor to account for the greater likelihood of unemployment among renters. This random assignment allowed more detailed but still representative analysis of the characteristics of renters estimated to be unemployed following the start of COVID-19-related health orders. There are several key limitations with the approach used, including:  The analysis only includes data on job losses from April to June, 2020. Given continuing improvement in the local labor market since June, it is possible that some of the households with job loss may have returned to work. However, given the lack of additional relief, it is likely any rental debt incurred would still need to be paid back. Additionally, the analysis only includes households that experienced unemployment and does not include households who experienced a reduction in income due to reduced hours or furlough.  Assumption used to determine the extent of employment among renter households. The Terner Center analysis uses data from a 2018 survey in order to estimate the extent of unemployment among renter households. It is possible that the actual level of unemployment among renters due to COVID-19 is higher or lower than what was found in 2018.  Unemployment Insurance and other federal stimulus were not included in the Terner Center’s assessment of the ability to pay rent. The approach used does not account for unemployment insurance (UI) benefits (conventional or supplemental CARES Act) or federal stimulus payments given limited data on this. However, given the limited and onetime nature of these payments, it is possible that the ability to meet future rent payments may be limited. Additionally, other studies have estimated that UI take-up is around 65 percent, indicating that a nontrivial share of unemployed individuals may not be receiving UI benefits.21 To account for this limitation, and as with our low end estimate, we adjusted the maximum unpaid rent amount by 50 percent to account for unemployed renters paying some or all of their rent through some combination of savings, loans, and federal and state support from unemployment insurance and stimulus payments. 21 Urban Institute, “How Much Assistance is Needed to Support Renters through the COVID-19 Crisis?” (Brief published June 15, 2020). Budget and Legislative Analyst 16 Memo to Supervisor Preston October 27, 2020 Exhibit 5: High End Estimate of Unpaid Rent Assumptions/ Inputs Total Number of Renter Households experiencing Job Loss Share of Total Renter Households in San Francisco Median Rent for Impacted Renter Households Total Maximum Rent Due Estimated Unpaid Rent (50% of maximum) Value 33,200 15% $1,970 $65,404,000 $32,702,000 Sources: BLA analysis based on Terner Center analysis and other sources. See Appendix B for full methodology. For illustrative purposes, based on our model and using Terner Center data, we estimated unpaid rent related to COVID-19 job loss under three other scenarios (Scenarios 2-4) based on information on three sub-groups imputed by the Terner Center. Estimates for these sub-groups were then modelled assuming they were unable to pay 50 percent of rent for 1, 3 or 6 months. The estimated amount of unpaid rent for these scenarios are set out in Exhibit 6 below. Given the limited amount of data on the extent of nonpayment across time, we have provided estimates for non-payment of 50 percent of rent for a best-, central-, and worst-case (i.e. 1, 3 and 6 months). Our three alternative scenarios all assume a smaller share of renter households have not paid some or all of their rent since sheltering-in-place began compared to our high estimate. This is because we assume that only the following sub-groups of the renter household population did not pay rent:  Scenario 2: renter households with incomes of 80 percent of Area Median Income or less.  Scenario 3: renter households already rent-burdened (i.e., paying more than 30 percent of their gross income on rent) before sheltering-in-place began.  Scenario 4: renter households that lost at least half of their household income after sheltering-in-place began. As can be seen in Exhibit 6, these sub-groups amount to between 4.7 and 7.5 percent of all renter households in San Francisco, or substantially less than our Scenario 1 assumed 15 percent share of all renter households estimated by the Terner Center as experiencing unemployment since the onset of the pandemic. Budget and Legislative Analyst 17 Memo to Supervisor Preston October 27, 2020 In Scenario 4, which focuses on unemployed renter households who are estimated to have lost at least half of household income, unpaid rent for 17,000 impacted households is estimated at $16.7 million for one month to $100.5 million for all six months, with a central estimate of $50.2 million for three months. Scenarios 2 and 3, which focus on low-income and rentburdened households, respectively, result in unpaid rent estimates of between $15.5 million to $92.9 million and $10.5 million to $63.1 million for one to three months of non-payment. Exhibit 6: Alternative Scenarios for Estimated Unpaid Rent Scenarios using Job-Loss Methodology Unpaid Rent Scenario Scenario 2: Started with Incomes of 80% of AMI or Less Scenario 3: Were Already Rent Burdened Pre-COVID-19 Scenario 4: Lost At Least Half of Household Income # of Renter HHs Impacted % Total Renter Households 1 month 3 months 15,700 6.9% $15,483,931 $46,451,794 $92,903,588 10,700 4.7% $10,524,181 $31,572,542 $63,145,084 17,000 7.5% $16,749,516 $50,248,549 $100,497,097 # of Months of Unpaid Rent (April - Sept) 6 months Source: BLA analysis based on Terner Center analysis and other sources. See Appendix A for full methodology. Notes: “Rent burdened” is defined as spending 30% or more of disposal income on gross rent. Estimates of Unpaid Rent by Building Size To provide further detail on the types of renters affected and the distribution of unpaid rent, we used an analysis of the Household Pulse Survey by the Terner Center to estimate the distribution of unpaid rent by building size.22 Exhibit 7 below provides a summary of these estimates for Scenario 1 (i.e. where all households impacted by job loss pay 50% of total rent due). A further breakdown of unpaid rent by more granular building size categories is provided in Appendix C. Based on this more granular analysis, renters in single family residences and buildings with less than 10 units could account for $20.7 million in unpaid rent per month in San Francisco County based on our high end estimates. 22 Terner Center for Housing Innovation, “The ongoing housing crisis: California renters still struggle to pay rent even as counties re-open” (Blog post dated October 2, 2020). Note the breakdowns provided in this analysis are at the San Francisco Metropolitan Statistical Area. However we have applied the MSA distribution to SF County as this is the most current data. It is possible the impact on single family residences are overstated given the larger share of these units in the SF MSA compared to San Francisco county (26% vs. 12%, respectively). Budget and Legislative Analyst 18 Memo to Supervisor Preston October 27, 2020 Exhibit 7: Estimated Unpaid Rent by Building Size for High Estimate Building Size Single Family 2-4 Units 5 to 19 Units 20 to 49 Units 50 or More Units Other3 Total % of All Renter Households (SF Only)1 Est. % of Renter Households Behind on Rent2 12% 22% 27% 13% 25% 0.3% 100% 34% 22% 15% 6% 16% 8% 100% # of Months of Unpaid Rent (April - Sept) 1 month 3 months 6 months $11,099,907 $7,168,832 $4,763,798 $1,828,645 $5,300,889 $2,539,928 $32,702,000 $33,299,722 $21,506,496 $14,291,395 $5,485,935 $15,902,668 $7,619,783 $98,106,000 $66,599,444 $43,012,992 $28,582,791 $10,971,870 $31,805,337 $15,239,565 $196,212,000 Sources: BLA Analysis based on Terner Center analysis of Household Pulse Survey for week 13 (Aug. 19 – Sept. 3); American Community Survey 2019 1-year estimates, Table B25066. Notes: 1 Breakdown by building size for all renter households is provided for San Francisco County for reference. 2 Breakdown by building size for those behind on rent is based on an analysis of the Household Pulse Survey by the Terner Center. However, data is for the San Francisco Metropolitan Statistical Area, which includes San Francisco, Alameda, Contra Costa and Marin counties. 3 Other includes mobile homes, RVs, boats, and other dwelling types. Given small sample sizes, estimates for this group are likely to be inaccurate. Another way in which the rental market has been affected by the pandemic and sheltering-in-place in addition to an increase in non- or partial payment of rents has been a reduction in demand for rental housing and/or an uptick in tenant requests for rent reductions. This is likely adding to a reduction in income for landlords beyond non-payment of rent. While tenants have been generally protected from eviction due to City orders adopted pertaining to non-payment of rent since sheltering-in-place, these orders require most tenants to pay their rent arrears in the future, meaning that landlords should eventually be paid for their lost income though they will have to manage cash flow problems if their expected income is deferred for six or twelve months (see Section 3 for details on eviction protections). Tenants vacating properties or asking for rent reductions, on the other hand, could mean permanently reduced income for landlords. These trends in reduced rental prices and increased vacancies are supported by data from San Francisco Apartment Association surveys of its members. While not representative of all landlords, the surveys show a significant and increasing share of landlords have received notices from tenants who wish to vacate their unit, break their lease, or are requesting rent reductions. Exhibit Budget and Legislative Analyst 19 Memo to Supervisor Preston October 27, 2020 8 below shows that from June to August 2020, the number of residential tenants submitting a Notice to Vacate to their landlords increased from 7.4 to 14 percent of tenants in properties managed by survey respondents. The number of housing providers who had received requests for rent reductions increased from 21 percent in May to 45 percent in August and almost 40 percent had a resident break their lease or give notice unexpectedly. Exhibit 8: San Francisco Apartment Association COVID-19 Survey Responses, April – August 2020 Survey Month (2020) Apr. May Survey Statistic Response Information # of Respondents # of Apartments Covered % of residential tenants who: Submitted Notice to vacate since Shelter in Place Currently vacant or will be vacant on 1st of the month % of respondents/ housing providers who: Received requests for rent reduction* Granted requests for rent reductions* Had resident break their lease or unexpectedly give notice June July Aug. 315 352 301 289 253 10,377 13,961 11,829 5,908 4,593 7.4% 13.0% 8.5% 14.0% 10.5% 21.0% 28.0% 21.0% 34.9% 26.2% 45.4% 35.1% 17.0% 24.0% 16.0% 39.8% Source: San Francisco Apartment Association, Survey on Non-Payment of Rent due to COVID-19 Notes: *Includes both temporary and permanent rend reductions. 3. Summary of Legal Protections and Financial Assistance Available for Tenants and Landlords In response to the economic impacts of the COVID-19 pandemic, local, state and federal governments have primarily focused on providing temporary legal protections from eviction and foreclosure due to non-payment. These protections focus on preventing evictions due to non-payment of rent (or mortgages) that was due during the pandemic. However, tenants’ liability for unpaid rent remains and landlords are authorized to collect this debt through civil court collection processes starting in March 2021.While eviction protections largely continue in San Francisco through March 31 2021, rent is due on-time and in full starting February 1st, 2021, with nonpayment subject to normal eviction and collection proceedings. Direct financial assistance has largely been provided through the expansion of unemployment insurance benefits by the CARES Act and subsequent Budget and Legislative Analyst 20 Memo to Supervisor Preston October 27, 2020 federal Executive Action. The continuation of these supplemental benefits beyond the end of the year is dependent on further federal or state action. Locally, the main program we identified was the primarily privately-funded Give2SF Housing Stabilization Program run by MOHCD since May. This program has committed $6.5 million of $7.5 million in total funding to provide financial assistance for 1,449 renter households. Continuation of this program is dependent on further funding which, to date, has been primarily from private sector donations. Local, State, & Federal COVID-19 Financial Assistance for Tenants and Landlords Direct financial assistance for tenants and landlords unable to make payments due to COVID-19-related financial hardship has been limited. Federal assistance has primarily come in the form of economic stimulus payments and expanded unemployment insurance benefits funded by the CARES Act, both of which have largely expired. While such assistance aims to replace a portion of income from involuntary unemployment, the expiration of supplemental UI benefits may lead to an increasing gap between household expenses (including rent) and income. Additionally, other than one-time Economic Impact Payments, we did not find any direct financial assistance programs available for landlords who may rely exclusively on rental income. Exhibit 9 provides a summary of federal, state and local assistance available. Exhibit 9: Summary of Financial Assistance and Rent Relief for Tenants affected by COVID-19 Source/ Authority Program Description/ Eligibility Period Covered Amount Federal Economic Impact Payment (Stimulus Payment) Non-dependent US citizens with adjusted gross income below limit ($150,000 for joint filers, $112,500 for Head of Household, $75,000 for individuals) One-time $1,200 / person CARES Act (Supplemental Unemployment Insurance) Tied to state unemployment Insurance benefit eligibility (see below) Mar 27 – Jul. 25 $600/ week ($2,598/ mo.) Lost Wages Assistance (Supplement UI) Supplemental Unemployment Insurance following CARES Act supplemental UI sunset. Aug. 1 – 29 $300/ week ($1,299/ mo.) Budget and Legislative Analyst 21 Memo to Supervisor Preston October 27, 2020 (continuation subject to funding) Pandemic Unemployment Assistance COVID-19-impacted workers who don’t qualify for regular UI benefits (i.e. independent contractors, business owners, self-employed workers or those with insufficient earnings or work history for UI). Feb 2 – Dec. 26 2020 State Unemployment Insurance Benefits Workers earning at least $1,300 during one quarter ($433/ mo.) in the 15 months preceding a claim. Ongoing Local MOHCD’s Give2SF Housing Stabilization Program (administered by community-based organizations under contract to MOHCD) Financial assistance program for SF residents facing financial hardship due to COVID-19related income loss or expenses. Based on funding ($7.5m available; $6.5m committed) $167 - $450/ week ($723 $1,948/ mo.) $40 - $450/ week ($173 $1,948/ mo.)* $4,500 / renter (avg.) ($3,000 $10,000) Source: BLA review of available assistance Notes: *Actual amount depends on wages earned in the highest earning quarter in the year preceding a claim. Local, State, & Federal COVID-19 Legal Protections for Tenants and Landlords Exhibit 10 below summarizes the various local, state and federal legal protections for tenants and landlords facing financial hardship due to COVID19. While these provide comprehensive protections from eviction for tenants behind on rent due to COVID-19-related hardship, they do not address outstanding rental arrears debt which can be collected as soon as March 1st, 2021 under state law (AB 3088). Landlords with mortgages face similar risks in terms of foreclosures due to non-payment. The CARES Act provides homeowners with federally-backed mortgages the ability to request up to 360 days in mortgage forbearance and mortgage loan servicers are prohibited from beginning foreclosure proceedings until January 1st, 2021. However, these protections still require deferred mortgage payments to be paid back. Additionally, homeowners with non-federally-backed mortgages, common for larger properties that Budget and Legislative Analyst 22 Memo to Supervisor Preston October 27, 2020 exceed federal mortgage limits23, are not afforded these protections and must rely on the discretion of their mortgage loan servicers for forbearance. Exhibit 10: Summary of Local, State and Federal Eviction and Foreclosure Protections 2020 Source/ Authority Mar 2021 Apr May Jun Jul Aug Sept Oct Nov Local – Mayor’s Eviction Moratorium (non-payment due to COVID) Mayor’s Eviction Moratorium: temporarily bans eviction for non-payment due to COVID-19 related financial impacts during the moratorium (3/13 – 9/30) until 6-months after the order ends (3/31/21). Provides 6-month extension for rent payments due during Eviction Moratorium. Local – Mayor’s Eviction Moratorium (all evictions) Mayor’s Eviction Moratorium: temporarily bans all evictions within 60 days of expiration of the order (currently 11/29/20), except evictions due to violence, health & safety or Ellis Act. Local – Board Legislation (Permanent Non-Payment Eviction Ban) Board Amendments to Rent Ordinance (7/27/20): permanent ban on eviction for non-payment during Governor’s Executive Order on evictions (3/16 – 9/30).* Can act as defense if other protections don’t apply. Does not prevent civil action to collect unpaid rent. Federal – CDC Eviction Moratorium Federal – CARES Act Relief for Homeowners with Federally-backed Mortgages Feb AB 3088 COVID-19 Tenant Relief Act: Eviction protection for non-payment from 3/1 – 8/31 if declaration provided. Unpaid rent can be collected in Small Claims court starting 3/1/21. AB 3088 COVID-19 Tenant Relief Act: Eviction protection for non-payment of rent from 9/1 – 1/31/21, If declaration provided + 25% of total rent paid by end of period. If 25% not paid by 1/31, eviction possible starting 2/1/21. State – Legislature: AB3088 (Deferral for future rent: 9/1 – 1/31/21) State – Court Emergency Rule 1 Jan Mar Board Amendments to Rent Ordinance (10/6/20): temporary ban on no-fault evictions starting 9/15/20 until 3/31/20. Local – Board Legislation (No-Fault Eviction Ban) State – Legislature: AB3088 (Past missed rent: 3/1 – 8/31) Dec Judicial Council Emergency Rule 1, prohibited issuance of summons or entering defaults for unlawful detainer actions (evictions). Ended Sept. 1 CDC Eviction Moratorium for non-payment of rent related to COVID-19 (until Dec. 2020). Requires: (1) written notice/ certification by tenant to landlord; and (2) Individuals earning <$99k ($198k married). Unpaid rent can be collected on in full after moratorium ends. CARES Act Mortgage Forbearance and Foreclosure Moratorium: The CARES Act allows homeowners with federally-backed loans (i.e. FHA, VA, USDA/ Fannie Mae, Freddie Mac) the option to request mortgage forbearance (i.e. deferment) for up to 360 days (two 180-day options) for COVID-19 related financial hardship. Federallybacked mortgage loan servicers are also prohibited from foreclosing on properties until Dec. 31. Any mortgage payments suspended under forbearance must be repaid. Source: BLA review of available protections. Note: *On September 23 2020, the Governor of California issued a partial extension to the Eviction Executive Order for commercial tenancies. The impact of this extension on the coverage period for residential evictions is currently an unsettled legal question. 23 2020 Federal Housing Finance Agency limits for mortgages in San Francisco county are $765,000 for one-unit, $980,325 for two-units, $1,184,925 for three-units, and $1,472,550 for four-units. Budget and Legislative Analyst 23 Apr Memo to Supervisor Preston October 27, 2020 APPENDIX A Detailed Tables with Non-Payment of Rent Estimates The tables below provide additional detail on the sources supporting the statistics presented above. Exhibit A1: Percent of Late Rent Payments by Month for National Multifamily Housing Council-surveyed Housing Providers, April - September 2020 vs April September 2019 % of Rent Month 2019 2020 Difference (Year over Year) April May June July August September 2.3% 3.4% 4.0% 3.4% 4.2% 4.5% 5.4% 4.9% 4.1% 4.3% 5.5% 5.4% 3.1% 1.5% 0.1% 0.9% 1.3% 0.9% Average (Apr – Sept) 3.6% 4.9% 1.3% Source: National Multifamily Housing Council data for 11.1 - 11.5 million apartment units each month. Notes: The percent of late payments are based on payments made by the end of the month and excludes any partial payments made. Data above excludes non-professionally managed rental properties and housing providers with five or fewer rental units. Budget and Legislative Analyst 24 Memo to Supervisor Preston October 27, 2020 Exhibit A2: Terner Center Estimates of Renter Households Impacted by COVID-19 Job Loss for the Top 10 Most Impacted Counties, April – June 2020 % of Impacted Renter HHs That: Median Rent of Impacted HHs Have Residents of Color Started with Incomes of 80% of AMI or Less Total Now Rent Burdened Lost At Least 50% of HH Income County Los Angeles 1,807,000 286,800 16% $1,500 78% 57% 74% 54% San Diego 530,500 79,800 15% $1,640 68% 56% 75% 51% Orange 443,900 68,500 15% $1,800 72% 55% 75% 47% Alameda 270,700 40,700 15% $1,750 78% 51% 70% 50% San Bernardino 260,800 37,800 14% $1,340 81% 47% 75% 54% Riverside 250,000 36,400 15% $1,410 76% 43% 71% 53% Santa Clara 278,100 33,500 12% $2,200 79% 47% 66% 44% San Francisco 226,000 33,200 15% $1,970 66% 47% 59% 51% Sacramento 238,900 33,000 14% $1,250 70% 56% 71% 55% Contra Costa 135,500 23,700 18% $1,830 73% 53% 71% 49% 5,943,000 903,000 15% $1,530 73% 53% 72% 52% California Renter HHs Impacted by Job Loss % Impacted by Job Loss Total Renter Households (HHs) Source: UC Berkeley Terner Center for Housing Innovation, Notes: Estimates of renter households are modelled based on Terner Center analysis of the 2018 American Community Survey 5-Year PUMS data and unemployment statistics from the U.S. Bureau of Labor Statistics Current Employment Survey. Exhibit A3: San Francisco Apartment Association COVID-19 Survey Responses on Non-Payment of Rent, April – August 2020 Survey Month (2020) Apr. May 315 352 Survey Statistic # of Respondents # of Apartments Covered % of residential tenants who: Did not pay ANY rent for the month due to COVID Paid partial rent for the month Total % residential tenants paying some or no rent June 301 July 289 Aug. 253 10,377 13,961 11,829 5,908 4,593 5.7% 3.2% 5.7% 3.2% 2.3% 2.5% 4.8% 3.1% 3.4% 6.5% 2.2% 4.0% 6.2% Source: San Francisco Apartment Association, Survey on Non-Payment of Rent due to COVID-19 Budget and Legislative Analyst 25 Memo to Supervisor Preston October 27, 2020 APPENDIX B Detailed Explanation of Methodology for Calculating Unpaid Residential Rent Estimates In order to estimate unpaid residential rent due to non-payment in San Francisco County, we used the following steps: Step Calculation Assumptions 1 Start with Terner Center estimates of renter households experiencing job loss for San Francisco County (Scenario 1) and relevant sub-groups (Scenarios 2 – 4) - Assumes Terner Center estimates using ACS & CES data are accurate and representative of actual SF renter household population in number and characteristics - Assumes job loss is strong and accurate predictor of non-payment of rent 2 3 Multiply Terner Center estimates of impacted renter households times the Terner Center estimate of median gross rent of impacted households from ACS 2018 - Assumes median rent from ACS 2018 is still accurate and representative Multiply estimates from Step 2 by percent of rent paid in each month (50%) and then the number of months of missed rent (1, 3 and 6 months) - Assumes 50% rent payment is similar to actual rent payment trends for impacted renters (on average) - Assumes the distribution of rent estimated from Terner Center is similar to that of actual renter households affected by COVID-19 job loss - Assume 1, 3 and 6 missed payment month inputs are similar (on average) to actual rent payment trends for impacted renters Budget and Legislative Analyst 26 Memo to Supervisor Preston October 27, 2020 APPENDIX C Detailed Breakdown of Unpaid Estimates by Building Size and Scenario The exhibits below provide a further breakdown of estimated unpaid rent by building size, including estimates for 5 – 9 units and 10 – 19 units, as distinguished in the American Community Survey (ACS). These estimates should be treated with caution as they assume the distribution of unpaid rent among these two groupings to be the same as the distribution of renter households among these two groupings in the ACS. According to the ACS 2019 1-year estimates, there were 32,955 renter household units in buildings with 5 – 9 units and 29,325 renter household units in buildings with 10 – 19 units, i.e. a ratio of 53 percent to 47 percent, respectively, for the 5 – 19 unit building category overall. This ratio is then used to distribute the estimated unpaid rent estimates for the 5 – 19 unit building size category across the two sub-categories (i.e., 5 – 9 units and 10 – 19 units). Exhibit C1: Unpaid Rent Estimates by Building Size for High Estimate (All Renters Impacted by Job Loss pay 50% of rent due) # of Months of Unpaid Rent (April - Sept) Building Size % of Rented HHs Est. % of Renter HHs Impacted by Job Loss 1 month 3 months 6 months Single Family Home 12% 34% $11,099,907 $33,299,722 $66,599,444 2 - 4 Units 22% 22% $7,168,832 $21,506,496 $43,012,992 5 - 9 Units 14% 8% $2,520,729 $7,562,186 $15,124,372 10 - 19 Units 13% 7% $2,243,070 $6,729,210 $13,458,419 20 - 49 Units 13% 6% $1,828,645 $5,485,935 $10,971,870 50+ Units 25% 16% $5,300,889 $15,902,668 $31,805,337 Other* 0.3% 8% $2,539,928 $7,619,783 $15,239,565 Total 100% 100% $32,702,000 $98,106,000 $196,212,000 Source: BLA Analysis based on Terner Center analysis of Household Pulse Survey for week 13 (Aug. 19 – Sept. 3); American Community Survey 2019 1-year estimates, Table B25066. Notes: Breakdown by building size is based on data for the San Francisco Metropolitan Statistical Area, which includes San Francisco, Alameda, Contra Costa and Marin counties. * Other includes mobile homes, RVs, boats, and other dwelling types. Given small sample sizes, estimates for this group are likely to be inaccurate. Budget and Legislative Analyst 27 Memo to Supervisor Preston October 27, 2020 Exhibit C2: Unpaid Rent Estimates by Building Size for Scenario 4 (Renters Impacted by Job Loss who experienced loss of at least half of household income and paid 50% of rent due) # of Months of Unpaid Rent (April - Sept) Building Size % of Rented HHs Est. % of Renter HHs Impacted by Job Loss 1 month 3 months 6 months Single Family Home 12% 34% $5,685,220 $17,055,661 $34,111,323 2 - 4 Units 22% 22% $3,671,778 $11,015,333 $22,030,665 5 - 9 Units 14% 8% $1,291,083 $3,873,248 $7,746,496 10 - 19 Units 13% 7% $1,148,870 $3,446,609 $6,893,218 20 - 49 Units 13% 6% $936,607 $2,809,821 $5,619,642 50+ Units 25% 16% $2,715,043 $8,145,129 $16,290,258 Other* 0.3% 8% $1,300,916 $3,902,748 $7,805,496 Total 100% 100% $16,749,516 $50,248,549 $100,497,097 Source: BLA Analysis based on Terner Center analysis of Household Pulse Survey for week 13 (Aug. 19 – Sept. 3); American Community Survey 2019 1-year estimates, Table B25066. Notes: Breakdown by building size is based on data for the San Francisco Metropolitan Statistical Area, which includes San Francisco, Alameda, Contra Costa and Marin counties. * Other includes mobile homes, RVs, boats, and other dwelling types. Given small sample sizes, estimates for this group are likely to be inaccurate. Budget and Legislative Analyst 28