Why Does San Francisco Have So Few Families?

Comparing San Francisco to Other U.S. Cities
Article
May 1, 2006

 

It’s not news that San Francisco has proportionally fewer children than almost any other city in the country; only 14.5 percent of San Francisco’s population in the 2000 census was under the age of 18, compared to 25.7 percent nationwide. Moreover, we seem to be separating ourselves further from the pack, with children declining as a percent of the population in every census since 1960. This is not a pleasant fact for those of us who care about San Francisco, believe a dense, diverse, exciting urban environment benefits from the presence of families and young people, who have confidence a city can be a great place to raise kids.

During the last few years, policy makers, civic leaders, and the news media have been clamoring to understand why San Francisco has so few families, and what can be done to reverse the trend. Understandably, most assessments of the problem point to the obvious culprits, notably the city’s high cost of housing and a lack of trust in the local public school system. Families, according to the most common version of the story, want to stay in San Francisco, but are being forced to pack up and leave by high housing costs and the lure of higher quality schools in other locales. Certainly there is some truth to this story for many families.

But could there also be something else at work? When the 2000 census was taken, 19.37 percent of San Francisco households had at least one person under 18—the lowest rate of any major city in the nation. But in San Jose, a geographic neighbor and one of the few cities where housing prices match San Francisco’s, 43.05 percent of households included a child—an increase of 0.73 percent during the 1990s, when housing prices skyrocketed.

How much can we expect local policy to impact the number of families that live in San Francisco? Housing costs and schools matter, but could it also be that what’s happening in San Francisco is part of a larger nationwide trend that isn’t as simple as higher-priced housing? Is there something else about San Francisco that keeps families away? We don’t have the answers to these questions, but we think a closer look at some comparative data from San Francisco and other U.S. cities helps to clarify what’s happening and refine the policy questions around this issue.

 

Families and Cities: San Francisco in Comparison

What trends can we observe about families not just in San Francisco, but in cities generally? The table in Appendix A compares the 50 largest cities in the United States (central cities only—not the entire metro area). There is a huge variation between urban areas in the percentage of family households, as well as the degree of change that occurred over a decade.

It is worth noting that, despite the conventional wisdom and the frequently cited figures above, San Francisco is not exactly losing families at an astounding rate. In fact, in 1990 San Francisco had 63,926 households with a person under 18, while in 2000 that figure was 63,867—a difference of only 59 households! Rather, the population grew as a whole, but family households simply declined as a percent of households in the city, contributing to the perception of fleeing families. San Francisco’s child population, in sheer numbers, has declined somewhat since the 1980s, but at a slower rate than during the period of nationwide suburbanization during the 1960s and 70s.

Looking at the big picture, one of the most dominant nationwide urban demographic trends during the last 20 years is the migration of Americans to the lower-cost, low-density sun belt communities of the South and Southwest. Consider the cities with a population of 100,000 or more that had the fastest population growth between 2003 and 2004: Port St. Lucie, Florida; Elk Grove, California; North Las Vegas, Nevada; and Gilbert, Arizona. Phoenix added nearly 30,000 residents in that year alone, and Las Vegas nearly 20,000—changes of a magnitude San Francisco sees over a decade.

Similarly, many of the cities that have both a high percentage of households with kids and saw growth in this percentage during the 1990s are in these population growth regions: Fresno (44.76 percent of households with a person under 18, 2.97 percent growth in those households during the 1990s); Phoenix (39.74 percent of households, 2.98 percent growth); Fort Worth (39.15 percent of households, 2.84 percent growth); Long Beach (38.84 percent of households; 5.69 percent growth); Las Vegas (35.67 percent of households, 1.53 percent growth).

Alternatively, many of the cities that are economically comparable to San Francisco—places with higher population densities, highly educated populations, knowledge-based economies, and high wage/cost structures—are primarily on the low end of the scale in terms of number of households with a person under 18, and most lost ground or stayed nearly static in this area during the 1990s. These cities include Austin (29.67 percent of households, -.052 percent growth); Portland (27.01 percent of households, -0.45 percent growth); Boston (25.65 percent of households, 0.45 percent growth); and Seattle (19.37 percent of households; -0.85 percent growth).

Aside from the high housing costs and low public school enrollment, many of the things that make San Francisco a uniquely urban place, and many of the things residents love about the city, are also things that are correlated with lower numbers of families. These include the city’s density, its reliance on public transit, its universities and access to education. This does not mean urban places are somehow inherently anti-family. But it may mean that given national demographic trends, we may need to adjust our expectations about how many families would choose to live here—even if the substantial housing and schools issues were addressed.

 

Which Families Are Leaving San Francisco and Which Ones Are Staying?

To look closer at what types of families are staying in or leaving San Francisco, we looked at individual-level census data for families who lived in San Francisco in 1995, and compared families who moved away from San Francisco between 1995 and 2000 (the most recent years for which this type of data are available) with those who didn’t. We take a similar look at families who lived in all central cities in the U.S. in 1995 for comparison.

Most of the figures in the table in Appendix B are simply averages, which means there are many families who fall above or below these levels. However, this information does help paint a picture of the general characteristics of the “typical” family who left the city. While low-incomes families clearly face the most pressure from the housing market to relocate away from San Francisco, these data do not suggest that the typical family leaving San Francisco is a low-income family forced out by high housing costs. In fact, families who left and those who stayed look surprisingly demographically similar and families who left actually had slightly higher household incomes than those who stayed. Families who stayed and left are both very well educated by national standards, and have kids that are roughly the same ages. It’s worth noting, however, is that people who stayed in San Francisco were spending more on housing—on average, about $150 more per month on mortgage payments, or $70 more per month in rent.

When people move away from San Francisco, where do they go? The table on page 21 shows the most common destinations for people who move away from San Francisco. While some are leaving for other states, primarily in the West and South, the vast majority are simply moving to Bay Area suburbs. This is not particularly surprising, as it is a well-documented trend that has affected American cities for the past 50 years.

As one can see in the table in Appendix B, it is seems that many families leaving San Francisco have the means to stay, but are choosing to migrate to the suburbs where they can have a bigger home (3.42 bedrooms on average for those who left, just 2.99 for those who stayed) in a low-density environment (about half of those who left found housing in a single family building, compared to one-quarter who stayed). They are also becoming homeowners after they leave, a reflection of the difficult market in the city.

Another difference that stands out between people who left and stayed in San Francisco is the level of enrollment in public schools. About 80 percent of those kids who stayed were in public schools, compared with 85 percent who left. This suggests that parents are less than thrilled about San Francisco schools. Low-income parents who distrust the public schools available have no option but to change their place of residence if they wish to change their children’s schools. For some middle- and higher-income families, the cost of private schooling combined with the marginally higher housing costs in San Francisco could easily be a deciding factor about whether to stay in the city. As discussed above, public school enrollment rates are an indicator of the number of families in an urban area, and San Francisco is unique among large U.S. cities in its low enrollment in public schools.


Principles Going Forward

Given all of these data, what does it actually tell us about which types of policy actions can be effective at the local level? If we want to increase the number families in the city as a percent of the population, how should we proceed?

 

1. CLARIFY THE POLICY GOAL.

Much of the discussion that has occurred to date seems to conflate the issue of families with more general economic issues. One goal that is widely shared is to provide low-income families who wish to stay here with the economic means to do so. But if the goal is to increase the number of children who live in the city, regardless of income level, a different set of policy issues come into play. Even if we found a way to ensure all low-income families have the means to remain in San Francisco if they choose, the city would still remain near the bottom in the United States in terms of the number of families.

 

2. TAILOR POLICY APPROACHES FOR DIFFERENT TARGET GROUPS.

Policies aimed at keeping families in the city that are primarily economic in nature will surely benefit low-income families. For instance, creating more family-friendly affordable housing will provide some families with the opportunity to remain in the city. But subsidy-based approaches make little sense for retaining the large numbers of families with household incomes near or above the median. For example, it makes little financial or policy sense to use limited affordable-housing funds to target high-income families, or to use public funds to subsidize their child care. Policy Policies aimed at keeping families in the city that are primarily economic in nature will surely benefit low-income families. For instance, creating more family-friendly affordable housing will provide some families with the opportunity to remain in the city. But subsidy-based approaches make little sense for retaining the large numbers of families with household incomes near or above the median. For example, it makes little financial or policy sense to use limited affordable-housing funds to target high-income families, or to use public funds to subsidize their child care. Policy solutions aimed at encouraging these families to stay in or come to San Francisco must be structured differently. For example, to offset the cost of housing for these families, it is necessary to pursue policies that decrease the average cost of market-rate housing. The only practical way to accomplish this is to increase the overall supply of housing. Similarly, changes that increase parents’ confidence in public schools can have a broad-based effect of making families at all income levels more likely to stay.

 

3. BE REALISTIC ABOUT WHAT IS AND IS NOT WITHIN THE CONTROL OF LOCAL GOVERNMENT.

Part of the reason for San Francisco’s low and declining proportion of families has to do with macro-level demographic and economic shifts, and people’s housing preferences generally. The reason San Francisco has so few families cannot be reduced to the explanation families are being “forced out.” Based on the data, it is probably unrealistic for San Francisco to expect to have as many families as cities like Phoenix, at least not for the foreseeable future, and if some families prefer suburban communities to a city, there’s little we can do to stop them. But it is certainly realistic to create an environment in which families at least believe San Francisco is an option, and are not forced to leave the city despite a strong desire to stay.

For all of us who love great cities and the cultural, social, environmental and experiential benefits they provide, it is worthwhile devoting energy to providing young people the opportunity to share them. 

 

appendix A.

Families In Urban Centers: Comparing the Number of Households with Children in the 50 Largest Cities in the United States

 % of Households with One or More People Under 18 in 2000Change Between 1990 & 2000
El Paso, Texas48.64%-1.45%
Fresno, California44.76%2.97%
San Jose, Californa43.05%0.73%
Virginia Beach, Virginia42.06%-2.72%
Detroit, Michigan41.51%1.70%
Arlington, Texas40.96%1.12%
San Antonio, Texas40.94%-0.52%
Phoenix, Arizona39.74%2.98%
Fort Worth, Texas39.15%2.84%
Long Beach, California38.84%5.69%
Jacksonville, Florida37.89%-0.01%
Houston, Texas37.76%1.81%
Los Angeles, California37.55%2.74%
Memphis, Tennessee36.86%0.06%
Colorado Springs, Colorado36.47%-1.06%
Mesa, Arizona36.33%-0.64%
Las Vegas, Nevada35.67%1.53%
New Orleans, Louisiana35.32%-0.68%
Cleveland, Ohio35.05%0.65%
Dallas, Texas34.94%2.96%
Milwaukee, Wisconsin34.77%0.10%
Wichita, Kansas34.77%0.49%
Chicago, Illinois34.31%0.28%
Sacramento, California34.16%1.13%
Oklahoma City, Oklahoma34.01%-0.44%
Charlotte, North Carolina34.02%0.01%
New York, New York33.96%2.31%
Oakland, California33.50%0.59%
San Diego, California33.35%0.99%
Indianapolis, Indiana33.32%-1.27%
Albuquerque, New Mexico33.13%-1.26%
Philadelphia, Pennsylvania32.73%0.90%
Baltimore, Maryland32.55%-1.76%
Omaha, Nebraska32.47%-0.11%
Tucson, Arizona32.47%0.47%
Kansas City, Missouri31.68%-0.42%
Miami, Florida31.57%-0.83%
Tulsa, Oklahoma31.37%-0.19%
Columbus, Ohio30.95%-0.68%
St. Louis, Missouri30.08%-11.32%
Nashville-Davidson, Tennesee29.86%-2.18%
Austin, Texas29.67%-0.52%
Louisville, Kentucky29.45%-0.54%
Honolulu, Hawaii28.00%-0.63%
Atlanta, Georgia27.40%-3.88%
Portland, Oregon27.01%-0.45%
Denver, Colorado26.27%0.56%
Boston, Massachusetts25.65%0.45%
Washington, D.C.24.56%-0.94%
Seattle, Washington19.65%-0.85%
San Francisco, California19.37%-1.55%

Source: U.S. Census Bureau, Decennial Census 1990 and 2000, Summary Files 1 and 3.  Data downloaded here.

San Francisco ranks last among the 50 largest cities in the United States in the percentage of households with one or more person under 18. During the 1990s, this percentage dropped, although part of this decline can be attributed to a growth in the non-family population rather than a loss in the total number of families.

 

Appendix B.

Comparing Families Who Left SF To Those Who Stayed

 Lived in San Francisco in 1995Lived in Any Central City in 1995
 Stayed in SFLeft SFStayed in Same CityLeft
Mean House Value$445,629$362,304$241,307$263,764
Mean Payment on First Mortgage$1,576$1,421$1,157$1,417
Mean Gross Rent$1,047$977$709$767
Mean # of Bedrooms2.993.423.283.91
Percent In Single Family Detached Homes23.448.315.5356.5
Percent Homeowners36.44834.657.8
Mean Household Income$86,417$90,216$54,568$79,742
Mean Travel Time to Work (Minutes)30.2130.6236.6635.9
Percent with Bachelors Degree or Higher39.240.12037
Percent of Kids in Public School79.984.98587.7
Average Age of Children11.0310.1810.4310.4

Source: Census Bureau, Public Use Micro Sample 2000. Data downloaded from iPUMS, Minnesota Population Center, University of Minnesota. All differences between means are statistically significant at p=.01.

The table above compares the characteristics of families who left San Francisco between 1995 and 2000 with those who stayed. For comparison, it also includes similar data for all U.S. cities. In many respects, these families were very similar on average. However, notable differences include higher incomes, lower housing costs, and higher enrollment in public schools for those who left San Francisco.

 

Appendix C.

Where Do San Francisco Residents Move To and From?

Domestic Migration To and From San Francisco: 1995-2000

 Incoming Residents to SFOutgoing Residents to SFNet Outgoing SF Residents
TOP DESTINATION SITES   
Other California Counties75,766143,98068,214
Washington2,6225,1892,567
Nevada8242,9022,078
Arizona1,6462,533887
Texas3,5404,135595
Oregon2,1792,719540
Florida2,5563,006450
    
TOP STATES OF ORIGIN   
Connecticut1,580698-882
Ohio2,0051,042-963
District of Columbia1,822728-1,094
Pennsylvania2,5831,477-1,106
Illinois4,8112,192-2,619
New York10,0686,959-3,109
Massachusetts6,2612,961-3,300
    
TOP DESITNATIONS WITHIN CA   
San Mateo County14,12231,05216,930
Alameda County10,34126,64216,301
Contra Costa County4,95616,30611,350
Marin County3,1689,6176,449
Solano County9384,4703,532
San Joaquin County7963,9843,188
Sonoma County1,9834,3592,376
San Luis Obispo County9202,8961,976
Los Angeles County9,91011,8151,905
Sacramento County2,3303,9591,629
    
TOTAL DOMESTIC MIGRATION145,095203,29258,197

Source: U.S. Census Bureau, County-to-County Migration Flow Files.

The data above inclues only domestic migration – not international migration. Between 1995 and 2000, almost 60,000 more people left San Francisco for other U.S. locations than moved to San Francisco from elsewhere in the U.S.  In interstate migration, San Francisco saw a net loss of population to states in the West and South, including states such as Nevada, Arizona and Florida that have experienced extreme positive net migration from all across the country during the past decade. The city saw a net gain in population primarily from highly urbanized states on the East Coast. By far, most people who moved to or from San Francisco came from or went to another California county. Not surprisingly, many of the most common destinations for people leaving San Francisco were the surrounding Bay Area counties.

About the Authors: 

Greg Wagner is the director of SPUR’s good-government program. Nicole Lembo is an intern at SPUR, and a student at Gordon College in Wenham, Massachusetts.