The region’s housing inventory is beginning to expand again, but the pace of growth lags that of employment and population. Apart from the natural delay due to the time required to build a unit of housing, several characteristics of the post-recession recovery period contributed to this lag. Sharp drops in home prices and high rates of foreclosure made new single-family home construction difficult to achieve coming out of the recession. Stronger rents and job growth centered in the major urban centers led to a quicker recovery of multifamily permits, but shortfalls in both commercial and consumer financing slowed the pace at which permits were translated into units, leaving many entitled units unbuilt for years. Although housing stock has expanded at a slow pace, the location of new construction is shifting relative to earlier periods. Since 2010, housing production has concentrated in urban centers and suburban nodes close to transit stations and corridors. Now that employment levels are starting to reach previous peaks, current trends in prices, affordability, and overcrowding indicate that the region is struggling to provide housing to meet the needs of the growing workforce.
The California Department of Finance (DOF) estimates the Bay Area added 38,300 housing units between April 2010 and January 2014. The annual average number of units added during this period (9,600 units per year) is low compared to previous decades, likely in part because of effects of the Great Recession which led to an overhang of available housing and tighter credit markets. From 2000 to 2010, the Bay Area added 231,600 new housing units, for an average of 23,200 per year. During the 1990s, the region averaged 18,700 units per year, or 187,500 for the decade.
Santa Clara, Contra Costa, and Alameda counties produced the most new units in the region since 2010, as they did in the previous two decades. (Figure 4.1). Over time, Bay Area housing growth has increasingly concentrated in a smaller number of jurisdictions, with San Jose and San Francisco specifically taking on a larger share of the region’s growth. Between 2010 and the end of 2013, the five jurisdictions with the most new units (San Jose, San Francisco, Dublin, San Ramon, and Sunnyvale) accounted for 51 percent of the total growth, with San Jose and San Francisco together accommodating 37 percent of the total. During the 2000s, 40 percent of the region’s new units were added by just five jurisdictions (San Francisco, San Jose, Oakland, Santa Rosa, Brentwood) with San Francisco and San Jose alone accounting for 27 percent of the total. Similarly from 1990 to 2000, 35 percent of the region’s new units were added by five jurisdictions (San Francisco, San Jose, Santa Rosa, Antioch, and Fremont), with San Jose and San Francisco accounting for 22 percent of the total. The increasing concentration of units in San Francisco, the South Bay, and the junction of highways 680 and 580 shows housing growth concentrating near major job centers, rather than in more distant (historically less costly) suburban developments.
Although the average number of units has dropped overall, there has also been a significant shift in product type, as is shown in Figure 4.2. Santa Clara County, for example, is adding as many multifamily units per year currently as in the previous decade and far more than in the 1990s. At the same time, single family production has dropped by more than two thirds in the county. San Francisco has virtually come to a standstill on single family additions to stock, although even the “increase” in single-family stock in San Francisco may be an artifact of changing definitions rather than a real increase in units (see note to Figure 4.2b). In contrast, the city is adding more multifamily units per year than in the 1990s although fewer than in the 2000s.
Note: The change in number of units by type is obtained by comparing the stock of respectively single-family and multifamily units from one time period to the next. With each decennial census, the housing stock is fully enumerated and classified by type. While the overall counts are deemed very accurate, the type classification from decade do decade can vary substantially even for the same unit, possibly misrepresenting the distribution of housing unit growth between single family and multifamily units. For example, the chart implies more than 10,000 new single family units in San Francisco during the 2000’s. This is almost exclusively a classification artifact of the way some multifamily units were reclassified as single family units in the 2010 Census.
Most counties have seen declining vacancies since 2010, as shown in Figure 4.3. Only Napa and Marin counties had higher vacancy rates in 2013 than in 2005, although all vacancies are higher than in 2000, at the peak of the dot-com boom. Napa’s vacancy rate rose to 13.6 in 2013, far higher than in any of the other counties. Not surprisingly, with its economic strength, Santa Clara County had the lowest vacancy rate, below 4 percent. San Francisco’s relatively high vacancy rate is more surprising. This may be because the ACS measure of total vacancy rates includes seasonal units that are unoccupied or occupied for short temporary periods at the time of the survey, counting these as vacant relative to total stock (but excluding them when rental or homeowner housing alone is considered). This may also account for Napa’s relatively high vacancy rate over time.
For each county, vacancy rates are lower when considering either homeowner or rental properties alone. As shown in Figure 4.4, the difference is particularly evident for San Francisco and Marin counties, both of which have homeowner and rental vacancy rates close to Santa Clara’s very low rates.
While data from DOF provides a look at the types of homes built, housing permit data shows what type of growth is expected in the short term, as permits precede construction. Since the 1990s, per Figure 4.5, the proportion of multifamily units permitted in the Bay Area as a whole is trending upward. The dip in 2009 is likely recession related, with uncertainty and tight credit reducing permit levels.
The set of maps in Figure 4.6 shows the shift to multifamily permitting in long-term geographic perspective. Cities that predominantly issue single-family permits in gold and cities that predominantly issue multifamily permits in blue. The increase in the proportion of multi-family units since the 1990s is evident in both suburban and urban areas.
An alternative look at the data shows a few counties dominate this shift. Figure 4.7 highlights the increased concentration of housing growth in the counties closest to the Bay and the growing proportion of multifamily permits in a few other parts of the region. San Francisco, Santa Clara, and San Mateo counties are the only three counties that have not experienced sharp declines in the average number of permitted units in 2010-2013 compared to previous decades, with San Francisco being the only one that has experienced an increase over all time periods. The high shares of multifamily permit activity for all of the large counties in the region suggest that regional growth will continue to concentrate around existing job centers for the near future.
In order to understand the degree to which new home development has occurred in the region’s adopted priority development areas (PDAs)13, ABAG collected detailed information from local governments about the specific locations of housing developments permitted in 2013. As shown in Figures 4.8 and 4.9, nearly 70 percent of the 11,800 multifamily units14 permitted in the Bay Area in 2013 were in PDAs, while 2,600 units (22 percent) were outside of PDAs. ABAG was unable to ascertain the location of the remaining 1,000 units (9 percent). Santa Clara County had the highest number of permits issued for multifamily units within PDAs, followed by San Francisco and Alameda counties.
The more urbanized counties permitted a greater portion of their housing units within PDAs, with San Francisco permitting 94 percent of its housing units in PDAs, followed by Alameda County (83 percent) and San Mateo County (67 percent). Most of the jurisdictions in the areas closest to the Bay permitted a majority of their multifamily units in PDAs, and there were 11 jurisdictions that permitted more than 90 percent of their multifamily units within PDAs.
Most North Bay counties permitted the bulk of their multifamily units outside of PDAs, but their total numbers of permits are small compared to all permits in the region. Napa, Marin, and Solano15 permitted 100 percent of their reported multifamily units outside of PDAs. Sonoma County is the notable exception in the North Bay, permitting 64 percent of its multifamily units in PDAs with the city of Santa Rosa permitting 75 percent of its units in PDAs.
As part of the Regional Housing Need Allocation (RHNA) process, the California Department of Housing and Community Development (HCD) determines the total number of housing units expected to be needed in the Bay Area during the RHNA time period to accommodate projected jobs and population growth. The RHNA process then assigns each jurisdiction in the Bay Area responsibility for permitting a target number of housing units that will be affordable to residents across all income groups, broken into four affordability categories.16
California law requires units to be tracked by level of affordability or by the rents or mortgages that will be charged to families within different income categories. Generally a home is considered affordable if it costs no more than 30 percent of a household’s annual income. California divides households into four broad income categories each based upon the Area Median Income (AMI): very low income households (making between 0 to 50 percent of AMI), low income households (between 50 and 80 percent of AMI), moderate income households (making between 80 and 120 percent of AMI), and above moderate households (making more than 120 percent of AMI). Jurisdictions are required to track residential permits issued based upon each housing unit’s expected affordability level once completed.
An evaluation of the permits issued by local governments relative to their RHNA targets provides an assessment of how well jurisdictions are doing in meeting the region’s housing needs, particularly for affordable housing. The two most recent RHNA periods were 2007-2014 (RHNA 2007-2014) and 1999-2006 (RHNA 1999-2006). Between 2007 and 2013,17 Bay Area jurisdictions permitted 57 percent of the total need, compared to 92 percent for RHNA 1999-2006). The significant decline in permit activity during the 2007-2014 period is largely a result of the Great Recession.
As shown in Table 4.1, for RHNA 2007-2014, jurisdictions in Santa Clara County permitted the highest percentage of total need (74 percent) followed by San Francisco (64 percent) and Contra Costa County (62 percent). Solano County (38 percent) and Marin County (32 percent) permitted the lowest percentage of total need. This differs from the pattern in RHNA 1999-2006, where Contra Costa County permitted 138 percent of its total need, followed by Solano County (99 percent) and Sonoma County (94 percent). For RHNA 1999-2006, the county with the lowest percentage of need permitted was San Mateo, at 63 percent. The East Bay and North Bay counties that have seen particularly large drops in permitting relative to need are also places that were still working through the overhang of units in foreclosure left over from the housing bust in the early years of this decade.
|Very Low||Low||Moderate||Above Moderate||Total|
|RHNA||Permits Issued||Percent RHNA Met||RHNA||Permits Issued||Percent RHNA Met||RHNA||Permits Issued||Percent RHNA Met||RHNA||Permits Issued||Percent RHNA Met||RHNA||Permits Issued||Percent RHNA Met|
Source: ABAG survey of local jurisdictions
Figure 4.10 compares the percent of the total need in each income category for which local jurisdictions issued permits during both the 1999-2006 and 2007-2014 RHNA periods. For both time periods, jurisdictions were most successful in permitting market-rate units, particularly from 1999-2006, when the permits issued represented 153 percent of the need. Jurisdictions also struggled during both time periods to meet the need for affordable units, particularly those affordable to moderate income households. The ability of local jurisdictions to meet the need for affordable housing was also hampered by the dissolution of Redevelopment Agencies in 2011. Redevelopment agencies provided an estimated $250 million annually to the production of affordable housing in the Bay Area.19 It is likely that the impact of this loss of funding is not yet reflected in the permit data, since affordable units permitted until 2013 already had their funding in place.
The high share of market-rate units in the permits issued is also demonstrated in Figure 4.11, which compares the share of each income category for RHNA compared to the permits issued by jurisdictions, for both RHNA 1999-2006 and RHNA 2007-2014. The patterns are remarkably consistent for the two time periods, and show that Bay Area jurisdictions are permitting a greater proportion of units affordable to above moderate-income households relative to the need identified in the RHNA.
For RHNA 2007-2014, San Francisco permitted 45 percent of the need for units affordable to both very low- and low-income and 59 percent of its need for units affordable to very low-income households. Following San Francisco, jurisdictions in Sonoma County and Alameda County permitted 29 percent and 28 percent, respectively, of their need for units affordable to both very low- and low-income households.
For units affordable to above moderate-income households, jurisdictions in Santa Clara County permitted 139 percent of the allocation, followed by 109 percent in San Francisco, and 96 percent in Contra Costa County.
As a comparison, for RHNA 1999-2006 San Francisco stands out for permitting 80 percent of its need for units affordable to very low-income households, and 72 percent when considering units for very low- and low-income households. At 78 percent, jurisdictions in Santa Clara County permitted a higher percentage of very low- and low-income units. In every county in the region, jurisdictions exceeded the number of housing units needed for above moderate-income households.
Another challenge to meeting the need for affordable housing is the potential loss of existing units that have deed restrictions to ensure affordability. The California Housing Partnership Corporation (CHPC) has evaluated the deed-restricted affordable housing units in the Bay Area to assess their potential for converting to market-rate housing.20
Properties were considered to be at high or very high risk of conversion if they were owned by a for-profit or small non-profit developer and the US Department of Housing and Urban Development (HUD) or Low Income Housing Tax Credit (LIHTC) affordability restrictions were set to expire within the next one year (very high risk) or within the next five years (high risk). Table 4.2 shows the number of deed-restricted affordable housing units in the region and each county that are at risk of converting to market-rate rents within this time period, or those that likely have already converted.21
There are a total of 6,888 affordable housing units in the Bay Area that are at risk of conversion to market-rate housing in the next five years. Most of these units are located in San Francisco and Santa Clara counties. However, relative to other counties, San Mateo has a significantly higher proportion of its deed-restricted affordable units at high and very high risk of conversion.22
|High Risk||Very High Risk||Potentially Converted||Total High,
Very High Risk,
and Potentially Converted
Restricted Housing Stock
Source: ABAG from California Housing Partnership Corporation data
According to data from the Census and American Community Survey, from 2010 to 2013, the total owner-occupied units in the region decreased by 7,600 while the number of renter-occupied units increased by 52,700. As a result of these diverging trends, in 2013 there were 1,195,300 renter-occupied units in the Bay Area, which represented 45 percent of the total occupied housing units in the region.
This is the highest proportion of renter-occupied units for the region when compared to data for 1990, 2000, and 2010.
Although San Francisco still has the highest proportion of renter-occupied housing in the region, the proportion dropped from 65 percent to 64 percent between 2000 and 2010 and to 63 percent by 2013. Meanwhile, from 2010 to 2013, Solano County (12 percent) and Contra Costa County (9 percent) experienced the largest percentage increase in renter-occupied housing units. Both shifts could be explained by changing circumstances brought about by the Great Recession and its aftermath. In San Francisco, as the for-sale market softened, mortgage rates dropped, and rents held fairly steady, ownership became more affordable on the margin to the city’s employed residents. While these effects could also be part of the explanation for housing tenure trends in Solano and Contra Costa, they likely were counterbalanced by the degree of foreclosure and subsequent loss of ownership among residents. (Figure 4.11).
Both rental and sales data show the region’s housing market has recovered strongly, as indicated by rent and price increases.23
With few exceptions, asking rents, the current price someone would need to pay to sign a new lease in the current market, rose sharply throughout the Bay Area for 2010-2014. Rental data from RealFacts for housing complexes with at least 50 units shows that between 2010 and 2014, the average monthly rent in the nine-county Bay Area increased by 38 percent from $1,495 to $2,062 (Figure 4.13).24 The average monthly asking rent for the Bay Area in 2014 is 22 percent higher than the previous peak in 2001 of $1,689 per month. (Effects of such an increase on affordability are discussed in the next section: Mixed Message from Affordability Measures.
For 2010 to 2014, the counties with the greatest growth in employment in technology-related fields have experienced the largest percentage increases in average rents. Rents have increased by 44 percent in Santa Clara County, 43 percent in San Mateo County, 36 percent in San Francisco County, 34 percent in Alameda County and 33 percent in Marin County. In 2014, average monthly asking rents were highest in San Francisco County ($3,105), followed by San Mateo County ($2,367), Santa Clara County ($2,213) and Marin County ($2,204).
Table 4.3 shows the ten jurisdictions with the highest percentage change in average asking rents from 2010 to 2014, according to data from RealFacts. For this four-year period, there were 37 jurisdictions where average rents increased more than 30 percent and another 20 where average rents increase between 20 to 30 percent.
|1||Mountain View||Santa Clara||$889||58%|
|2||Los Altos||Santa Clara||$835||55%|
|5||Redwood City||San Mateo||$841||50%|
|7||Foster City||San Mateo||$839||48%|
|8||Palo Alto||Santa Clara||$949||45%|
|10||San Mateo||San Mateo||$762||45%|
Source: ABAG from RealFacts data. Not adjusted for inflation
As a complement to the RealFacts data, which represents average asking rents for leases and includes only data for developments with 50 or more units in the current rental market, ACS covers all building sizes and leases signed at any point in time and is thereby not so much an indicator of the current market. The two measures are indicative of different aspects of the market and are not strictly comparable.25 Their relationship is that today’s current market rental rates will shape the overall median rents as measured at a later point in time. As such, the change in current market rents is informative of the direction of the overall cost burden. With these caveats on comparability, median rents are significantly below the average rent calculated from RealFacts. According to ACS, the median gross monthly rent26 for the Bay Area increased by $134, or 10 percent from 2010 to 2013 (Figure 4.14).
Over this time period, based on ACS rental data, Santa Clara County experienced the largest increase in median gross monthly rent ($235), followed by San Mateo County ($222), and Alameda County ($137). This corresponds to a 17 percent change in Santa Clara, 15 percent in San Mateo, and 11 percent for Alameda. Table 4.4 shows the ten jurisdictions with the greatest increase in median gross monthly rent between 2010 and 2013. While cities with the largest increases shown from the RealFacts data are concentrated in San Mateo, Santa Clara and Marin counties, ACS data shows a wider impact of rising rents, with significant increases in dollar amounts and percent found in cities in the East Bay and North Bay as well.
|2||Palo Alto||Santa Clara||$293||17%|
|9||Unincorporated San Mateo County||San Mateo||$209||16%|
Source: ABAG from US Bureau of the Census American Community Survey 5-Year Estimates for 2010 and 2013
Between 2010 and 2014, the median sales price for homes sold in the nine-county Bay Area increased from $410,000 to $610,000—a 49 percent increase over the four-year period (Figure 4.15).27 Over this time period, Contra Costa County experienced the greatest percentage increase in median sales price (50 percent), followed by Alameda County (49 percent), and San Francisco County (46 percent). Even with these significant increases over the past four years, only San Francisco and San Mateo counties have exceeded their previous median sales price peaks.28 In 2014, San Francisco had the highest median sales price at $975,000, which was $207,000 more than its peak in 2007. San Mateo County had the second-highest median sales price in 2014 at $843,000 followed by Marin County at $836,000. Median sales prices in 2014 were more affordable in Solano County ($292,000), Sonoma County ($432,000), and Contra Costa County ($435,000).
In 2014, the five jurisdictions with the highest median sales prices were Atherton, Los Altos Hills, Hillsborough, Ross, and Portola Valley. The five jurisdictions with the lowest median sales prices in 2014 were Vallejo, San Pablo, Rio Vista, Suisun City, and Fairfield.
Table 4.5 lists the ten jurisdictions with the highest percent change in median sales price from 2010 to 2014. Many of the places on this list are “high recovery” cities—where the housing market had experienced a large number of foreclosures, but is now experiencing sales activity in a more normal range once again. Palo Alto is clearly not of that group, but is indicative of the city’s strong and changing market for housing.
Median Sales Price
|3||East Palo Alto||95%||$492,000||$240,000|
|7||Contra Costa County||66%||$457,000||$182,000|
Source: Multiple Listing Service Home Sales Records, calculations by ABAG
The median price is a helpful statistic to track the mix of housing over time and relative prices within the region. It is not equivalent to a price index for housing in the region. The Case-Shiller index for the San Francisco Bay Area is based on same-home sales and thus gives a more accurate picture of price trends for a similar quality home. It is also available for homes at different price levels. Figure 4.16 shows the Case-Shiller index for all homes (SF_CSAll), high priced homes (SF_CSHigh) and low priced homes (SF_CSLow) over an extended period of time. By this index, home prices have not yet fully recovered in the region from the peak of the housing bubble. Prices at the high end (the blue line) have indeed surpassed the earlier peak, but the previous peak for the lower priced home index was much higher, and relative prices for that segment of the market fell much further and have not recovered. The “recovery” of the median price of homes discussed earlier reflects not only price gains but more sales at the higher end of the market and fewer foreclosure sales.
While the sales price data provides a measure of what is happening in the current for-sale housing market, ACS provides data about changes in the monthly costs30 that all homeowners throughout the region pay for housing. Between 2010 and 2013, the median monthly costs for homeowners decreased by nine percent, from $2,217 to $2,015 per month. This decrease likely reflects some older homeowners paying off their mortgages, with other homeowners refinancing their existing mortgages to take advantage of historically low rates. As a result of this decrease, the regional median monthly costs for owners has only increased by $62 since 2005, reflecting both lower prices and reduced financing costs (Figure 4.17).
Contra Costa County experienced the greatest percentage decrease in median monthly owner costs (13 percent), followed by Sonoma County (11 percent), and San Mateo County (10 percent). For Contra Costa, Solano, and Sonoma counties, median monthly owner costs in 2013 in were lower than they were in 2005.
Table 4.6 shows the ten jurisdictions with the greatest increase in median monthly owner costs between 2010 and 2013. Many more of the places with the highest increases in owner costs are in expensive housing markets where prices have risen, but some North Bay cities in less expensive markets have also felt the effects. In contrast, Table 4.7 shows the ten jurisdictions with the greatest decrease in median monthly owner costs over this period. Several of these are in expensive housing markets, but many are in areas of Contra Costa and Solano counties that experienced a high proportion of foreclosures during the Great Recession.
|6||Palo Alto||Santa Clara||$382||15%|
|10||Unincorporated Solano County||Solano||$245||13%|
Source: ABAG from US Bureau of the Census American Community Survey 5-Year Estimates
|1||Portola Valley||San Mateo||-$611||-16%|
|2||Los Altos Hills||Santa Clara||-$592||-15%|
|8||Los Gatos||Santa Clara||-$331||-10%|
Source: ABAG from US Bureau of the Census American Community Survey 5-Year Estimates
Affordability can be measured in several different ways, depending on which population is being considered and the tenure of the household. Two measures, the housing wage measure of the National Low Income Housing Coalition (NLIHC) and the housing affordability index of the California Association of Realtors (CAR) focus on affordability relative to a transaction at a certain point in time. Measures using ACS data of the share of income households spend on housing provide a broader look at the overall welfare of households within a geographic area. The general picture given by these measures is of worsening affordability for renters and improved affordability for homeowners.
The National Low Income Housing Coalition (NLIHC) annually calculates the Housing Wage for communities across the United States in order to assess the affordability of rental housing. The Housing Wage is the hourly wage that a household must earn in order to afford a two-bedroom unit at the Fair Market Rent.31 The purpose of the Housing Wage is to demonstrate the discrepancy between the income needed to afford decent unsubsidized housing and the earnings available to many households.
In the Bay Area, Marin, San Francisco, and San Mateo counties had the highest Housing Wage in 2014 at $37.62 which corresponds to an annual gross income of $78,250.32 This was followed by Santa Clara County at $31.71 ($65,957 annually) and Alameda and Contra Costa counties, at $30.35 ($63,128 annually). This compares to the California Minimum Wage, which was $9.00 per hour as of July 1, 2014 ($18,720 annually).33 Although NLIHC cautions against comparing detailed Housing Wage data from one year to the next, Figure 4.18 shows the trends for each county. Although the housing wage dropped in several counties between 2010 and 2014, all of the region’s nine counties have housing wages well above the levels for 2005.
The California Association of Realtors (C.A.R.) produces a Traditional Housing Affordability Index (HAI)34 that measures the percentage of households that can afford to purchase the median priced home. Figure 4.19 shows the changes in the HAI for each county in the Bay Area and California from 1994-2014. With the exception of Solano County, the counties in the region are generally less affordable than the rest of the state. In 2014, Marin, San Francisco, and San Mateo were the least affordable counties in the region with only 14 percent of households able to afford a median-priced home. In contrast, 51 percent of households in Solano County could afford a median-priced home. Sonoma and Napa counties were also relatively more affordable since 29 percent and 25 percent of households, respectively, could afford a median-priced home. As prices have begun to recover, affordability has dropped sharply in all of the nine counties since 2012.
Housing costs are traditionally considered to be affordable when they are less than 30 percent of household income; households paying 30 percent to 50 percent of their income on housing are considered to be cost burdened, while households paying 50 percent or more of income on housing are considered to be severely cost burdened.35 Figure 4.20 shows the percentage of both owner and renter households paying 30 percent or more of their income on housing. In general, homeowner households have significantly higher incomes than renter households in the Bay Area. The median household income for owners has consistently been nearly double the median household income for renters.36 In 2013, the average regional median annual income was $104,000 for owners and $52,100 for renters. Nonetheless many homeowners are considered to be overpaying for housing.
Similar to the trends for household incomes, the situation has improved for owners while getting worse for renters. Between 2007 and 2013, the number of owner households that were cost burdened decreased by 22 percent, while the number that were severely cost burdened decreased by 34 percent. In contrast, the number of renter households that were cost burdened increased by 23 percent and the number that were severely cost burdened increased by 23 percent. In 2007, 43 percent of owner households (647,300) and 47 percent of renter households (474,000) had housing that was considered to be unaffordable. By 2013, 32 percent of owner households (469,400) and 49 percent of renter households (585,000) had housing that was considered to be unaffordable.
As shown in Figure 4.21, in 2013 Marin County had the highest proportion of cost burdened owner households (22 percent), while Napa and Sonoma tied for the highest proportion of severely cost burdened owner households (16 percent).
Figure 4.22 shows by census tracts in the Bay Area the proportion of owner-occupied households that are paying more than 30 percent of their income for housing, based on ACS five-year data for 2008-2012. The darker areas have the greatest proportion of cost burdened owner households.
Between 2007 and 2013, Solano County had the greatest percentage increase in renter households paying more than 30 percent of income on housing (31 percent additional households over 2007 levels), followed by Alameda County (28 percent), and Contra Costa County and Sonoma County (both 27 percent).37
As shown in Figure 4.23, in 2013, Solano County had the highest proportion of cost burdened renter households (31 percent) and the highest proportion of severely cost burdened renter households (32 percent). Figure 4.24 shows the census tracts in the Bay Area where renter-occupied households are paying more than 30 percent of their income for housing, based on ACS five-year data for 2008-2012. The darker areas have the greatest proportion of cost burdened renter households.
Housing overcrowding, defined by the US Department of Housing and Urban Development (HUD) as more than 1.01 occupants per room in a household, can deteriorate the quality of existing housing stock while also increasing the risk of spreading communicable disease.38 Overcrowding is not only an important consideration from the perspective of public health and the continued upkeep of quality housing stock, but is also an important feature of family life. Research shows crowding has deleterious effects on children’s well-being, with higher levels of stress, poorer personal health, and lesser performance in schools. Alarmingly, these effects persist into adulthood, emphasizing the importance of the measure.39
Renter households are more likely than owner-occupied households to experience overcrowding or severe overcrowding (when there are more than 1.51 persons per room). Six percent of all Bay Area households experienced overcrowding in 2013. However, 10 percent of all renter-occupied households were affected, while only 3 percent of owner-occupied households were overcrowded. (Figure 4.25.)
Historically, Bay Area counties have experienced overcrowding differently with San Mateo (7.8 percent of all households) and Santa Clara counties (6.9 percent of all households) generally having the highest rates of overcrowding. Conversely, Sonoma, Contra Costa, Solano and Marin counties have traditionally been the least overcrowded parts of the region. At a jurisdictional level, overcrowding is felt particularly strongly in jurisdictions in Santa Clara and San Mateo counties, as seen with the list of jurisdictions with the highest shares of overcrowding shown in Figure 4.26.
East Palo Alto stands out as the most overcrowded jurisdiction in the Bay Area with 29.3 percent of its occupied households considered overcrowded. Five other jurisdictions primarily in San Mateo and Santa Clara counties have overcrowding rates in excess of 10 percent. San Rafael, in Marin County, is the only North Bay jurisdiction to be among the top ten overcrowded jurisdictions with an overcrowding rate of 8.3 percent.
A key aspect of a Sustainable Communities Strategy is the reduction of greenhouse gas emissions by promoting a land use pattern that helps reduce vehicle miles traveled. One of the key strategies to accomplish this is encouraging housing growth in areas near jobs. This section looks at several simple indicators of the Bay Area’s balance between job and housing location.
The simplest indicator of the relationship between jobs and housing is the ratio of employment to the number of housing units. However, this ratio is not a simple scale with one end positive and, the other end negative. A better reading of the ratio is to use it to identify places that have few jobs relative to housing units (a low ratio relative to the average) as compared to places that have many jobs relative to housing units (a high ratio relative to the average), with the assumption that those communities with a close to average ratio have a better balance.
Figure 4.27 shows the jobs/housing ratio for the Bay Area and each of the region’s nine counties. The relative positions of the counties for the most part are stable, but the overall ratio clearly varies with economic conditions, dropping as employment dropped after 2008, and now rising in the locations where the economy is recovering most strongly. San Francisco and Santa Clara counties have the highest ratios, but Santa Clara’s has stayed much more stable as employment has recovered, while San Francisco’s has risen as housing production has lagged.
A closer look at the San Francisco case illustrates how quickly conditions can change for housing markets relative to jobs. Between 2000 and 2010, San Francisco added 29,000 people, the largest number on record for any ten-year period at the time, yet only 2,000 housing units, and even lost jobs. The county’s jobs/housing ratio was dropping in that period, and would have come closer to the regional average if the regional average had not also been dropping due to job loss. Rent levels were tempered during this period, although other aspects of the housing bubble delayed a downturn in housing prices for a few years. This improved situation for renters was short lived. Between 2010 and 2013, San Francisco added 32,000 people—more than in the entire decade between 2000 and 2010—while the housing stock grew by just one-sixth of that amount and jobs increased by over 70,000. As a result, the jobs/housing ratio for San Francisco is now higher than in 2002 and diverging from the regional average.
A measure developed by a group at UC Davis offers a more fine grained picture of jobs/housing balance by taking into account affordable housing available to low wage workers.40 Using the Longitudinal Employer Household Dynamics (LEHD) Origin Destination Employment Statistics Dataset, the measure compares affordable housing by area (as measured by reported housing expenditure in the American Community Survey) with low wage jobs in the workplace (as reported by LEHD). The jobs/housing fit is then the ratio of low wage jobs to affordable housing. A city with a lower ratio has more affordable housing available relative to low wage workers.41
Table 4.8 shows the cities with the lowest (less than 2) and highest (greater than 15) jobs/housing fit scores. The cities with the lowest scores are primarily places with both affordable housing and low income jobs, although in the case of Belvedere, the city has very few jobs. The cities with the highest scores are primarily high income residential areas with high priced homes (or, in the case of Colma, very little housing of any type) and a significant share of retail jobs in the employment mix.
|Cities with Jobs-Housing Fit Ratio < 2|
|East Palo Alto||0.98|
|Cities with Jobs-Housing Fit Ratio > 15|
Looking at the extensive geography of the region, many jurisdictions with the lowest scores are along the East Bay shore (Oakland, Berkeley, and Hercules). Many of the higher scoring places are along the West Bay shore, the South Bay shore, or the Interstate 680 corridor. Middle scores are clustered in the South Bay and parts of the North Bay. The distribution of scores gives some indication of where additional affordable housing investment would improve the jobs/housing fit.
A third indicator of jobs housing imbalance is the degree of commuting outside of a county. Figure 4.28 illustrates commute flows among counties. The size of the circle slice allocated to a county indicates the relative size of the number of working residents in a county (Alameda 693,000, Contra Costa 466,000, Marin 121,000, Napa 62,000, San Francisco 432,000, San Mateo 349,000, Santa Clara 817,000, Solano 184,000, Sonoma 226,000, as well as 162,000 who live beyond the Bay Area but work in the region). Although not directly shown in the figure, the largest job clusters are in Santa Clara County (916,000), Alameda County (700,000) and San Francisco (591,000). Most residents work within their county of residence, as shown by the arc contained entirely within each county slice. The lines between the slices indicate the flow of residents to workplaces. The lines represent flows of workers in both directions between counties, with the ends sized to indicate the total number of residents driving from the origin to the destination. Lines between counties are the color of the county sending the larger flow. For example, more residents from San Mateo County work in San Francisco than the reverse. For the most part, the largest between-county flows are with neighboring counties (Contra Costa to Alameda, San Mateo to San Francisco and Santa Clara). Santa Clara and San Francisco counties have the largest net inflows of workers in total from all locations, while Contra Costa and Solano counties have the largest net outflows of residents to work in other counties. The chart also shows flows into and out of the region-relatively insignificant compared to flows within the region, but the largest to Alameda and Santa Clara counties. The number of commuters into the region is larger than the than the number of Bay Area residents commuting out of the region. These commute flow patterns are a further indication of where vehicle miles traveled are generated and where further jobs housing balance could be sought.
13 These are locally designated infill areas with frequent transit service where a jurisdiction has decided to concentrate most of its housing and jobs growth for development in the foreseeable future.
14 ABAG focused its analysis on multifamily units to be consistent with the objective of encouraging higher density housing in PDAs and because there were fewer projects about which the location information was unavailable.
15 The location information was unknown for 32 percent of the units in Solano County.
16 For more information about the Regional Housing Need Allocation, see http://www.abag.ca.gov/planning/housingneeds/.
17 As of February 2015, data for the calendar year 2014 is incomplete.
18 The data for this report was compiled primarily from Annual Housing Element Progress Reports (APRs) filed by jurisdictions with the California Department of Housing and Community Development (HCD). In some cases data was compiled using planning documents generated by the jurisdictions (housing elements and permitting information sent to ABAG). APRs are the only source of information pertaining to the affordability levels of housing units permitted within a jurisdiction. Once the total number of housing units permitted is taken into account, some differences arise between APRs and two other commonly used sources for housing permitting data: the Construction Industry Research Board (CIRB) and the US Census. After a lengthy examination it was determined that CIRB and the Census gather data primarily through surveys of local planning and building staff whereas APRs are prepared directly by planning staff and sent to the state. The differences between these sources mostly arise due to who was specifically surveyed in a jurisdiction and whether or not the CIRB and Census researchers relied primarily on a jurisdiction’s website for permitting information. As no two data sources possess the same tally of housing units permitted for the Bay Area for the 2007-2014 period, APRs fall within an acceptable margin of error when compared to the Census or CIRB. All this data was sent to local jurisdiction staff for confirmation in December 2014.
19 Former Redevelopment Agencies were previously required to dedicate 20 percent of their tax increment money toward the development of affordable housing. At their peak the California Legislative Analyst’s Office estimates that California’s Redevelopment Agencies raised $5 billion annually with $1 billion devoted toward affordable housing development. Due to population size ABAG estimates the Bay Area’s share of that $1 billion to have been $250 million annually.
20 CHPC’s analysis assigned a level of risk for conversion based on the type of organization that owned the property, the existence of rental assistance contracts of insured/subsidized mortgages from the U.S. Department of Housing and Urban Development (HUD) that required affordability restrictions, and the existence of rent or mortgage restrictions through the use of Low Income Housing Tax Credits (LIHTC).
21 Properties where the deed-restriction has expired or will expire within the year, and no data is available about whether the deed-restriction has been renewed.
22 Marin County and Napa County do not have any units that are at high or very high risk of conversion, but Marin has 2,556 units and Napa has 1,463 units that are at very low or low risk of conversion.
23 Housing prices and rents are not adjusted for inflation because they are a key component within the CPI. As such, it would not be an appropriate to use the CPI (or the rental or housing component of the CPI) to adjust housing prices. Instead, we compare median price trends to trends in the Case-Shiller index which, as it is based on same home sales, shows cost relative to the same quality item. The section on affordability further compares the trend in home prices relative to overall income.
24 Calculations are based on data from RealFacts. Rents are not adjusted for inflation as explained above. Data includes only developments with 50 or more units. Not all jurisdictions are represented in the dataset.
25 In addition, the RealFacts data reports average rents while data from ACS reports medians. Typically for measures of cost, averages tend to be higher than medians as they are subject to skewing by the highest rents in the data. This will all other things equal exacerbate the difference between the two measures.
26 Gross rent is the amount of the contract rent plus the estimated average monthly cost of utilities (electricity, gas, and water and sewer) and fuels (oil, coal, kerosene, wood, etc.) if these are paid for by the renter (or paid for the renter by someone else). Gross rent is intended to eliminate differentials which result from varying practices with respect to the inclusion of utilities and fuels as part of the rental payment.
27 Data from Multiple Listing Service Home Sales Records for all available sales records from January, 1994 through November, 2014. Data is for sales identified as single family residences, townhouses, and coop/condo and excludes sales identified as multifamily, mobile homes, and vacant land. Sales prices are not adjusted for inflation.
28 The previous peak occurred in 2007 for Marin, San Francisco, San Mateo, and Santa Clara counties and in 2006 for Alameda, Contra Costa, Napa, Solano, and Sonoma counties.
29 With a total of at least ten sales between 2010 and 2014.
30 Selected Monthly Owner Costs are calculated from the sum of payment for mortgages, real estate taxes, various insurances, utilities, fuels, mobile home costs, and condominium fees. Listing the items separately improves accuracy and provides additional detail.
31 Fair Market Rents are defined by the US Department of Housing and Urban Development. The Housing Wage assumes that households pay no more than 30 percent of their income for housing. More details about the methodology for calculating the Housing Wage are available at: http://nlihc.org/oor.
32 Assuming full-time employment at 40 hours a week, 52 weeks per year.
33 http://www.dir.ca.gov/iwc/minimumwagehistory.htm. The California minimum wage will increase to $10.00 per hour on January 1, 2016.
34 The index is based on the median price of existing single-family homes sold and traditional assumptions about the costs of financing a home. Affordability is defined as a household paying no more than 30 percent of its income for housing. Data for Napa and Solano counties is only available going back to 2010. Details about the methodology for the HAI are available at: http://www.car.org/marketdata/data/haitraditional/.
35 When discussing the charts below, “cost burdened” and “severely cost burdened” are used as mutually exclusive categories, so the former refers to just the 30 to 50 percent segment, with the latter capturing households paying above 50 percent of household income on the costs of a home.
36 Source: American Community Survey 1-Year Estimates, calculated by ABAG. The average regional median income is the average of the median incomes for the nine counties in the Bay Area.
37 Solano County, while seeing a large increase in renter households that are cost burdened, conversely saw the largest decrease (35%) of cost burdened homeowner households. This may be two sides of the same coin—homeowner households paying lower interest rates for mortgages, displaced homeowners and unemployed renters with less income and higher rents.
38 US Department of Housing and Urban Development in its report “Measuring Overcrowding in Housing” chose to examine overcrowding through the “prevalence of communicable diseases in overcrowded environments and the effects they have on a child’s growth and development” p.2
39 See, e.g. Solari, C. D., & Mare, R. D. (2012). Housing crowding effects on children’s wellbeing. Social Science Research, 41(2), 464–76. doi:10.1016/j.ssresearch.2011.09.012
40 The methodology for this measure is further described in Chris Benner and Bidita Tithi "Jobs-Housing Fit in the Sacramento Region", 2011 UC Davis Center for Regional Change.
41 A map of Bay Area cities by jobs-housing fit score can be seen at http://mappingregionalchange.ucdavis.edu/node/553