The Great Bay Area Migration Survey

Background

I’ve heard a lot of people talking about moving out of SF or the Bay Area due to the pandemic, and I got curious to quantitatively understand the phenomenon and how real it is.  I recently created a survey to understand these migration patterns (or lack thereof) and am sharing the amalgamated results here.

Doing large-scale surveys effectively is tough; it’s hard to capture a wide range of responses into a fixed set of bins; if you have too few bins you lose valuable detail, and if you have too many you sometimes lose the big picture and end up with a data processing nightmare.  Separately, the wording of questions is extremely important and can strongly sway responses.  I did my best, and there are already several mistakes I regret, most notably not providing options for midwestern and southern cities as places to migrate to and not providing enough distinctions between large cities, small cities, and small towns.  Fortunately, there was a strong write-in contingent for a couple of locations, and I created categories for them.  

I asked anyone who lives in the Bay Area, works in the Bay Area, or considers most of their social network to be in the Bay Area to complete the survey.  In total I got 222 responses over 2 days on September 8th-10th, 2020.  The vast majority are from Facebook friends of mine, and a handful came in from friends-of-friends.  I do not expect these results to necessarily generalize to the Bay Area population as a whole, so these are mostly interesting to people in my social network to understand what the people I care the most about are up to.  If you have no idea who Matt Bell is, this data may be less useful to you. :-)  

Also of note -- the last 80 responses or so came in when SF looked like this.  While this doesn’t affect data on what migration has already happened, it may have affected people’s thinking about their future desire to live in the Bay Area.  

Dolores Park got too real even for the hipsters

To give you a sense of the demographics of the survey respondents, here’s some quick information about them:  The people who filled out the survey are mainly in their mid-20s to mid-40s.  About 61% of them are in a long term primary relationship, and 25% of them have kids.  Of the remaining 75% who don’t have kids, 45% want to have kids in the future.  The group is heavily skewed toward people who work in tech; in fact, people working jobs in the software industry comprise half of the entire total, and software + hardware + biotech together comprise 68%. There’s a long series of charts in the appendix that lets you dive into all the demographic information in more detail.  

On to what we have learned...

How has the pandemic impacted people?  

Work-from-home is here to stay, and partial work-from-home is likely our future post-pandemic

A lot of people are working from home.  This is probably the most massive shift in the nature of work we’ve seen in a long time.  In the early 2010s, companies like Yahoo banned work-from-home, and it’s generally been frowned upon as a practice.  Now companies don’t have a choice.

It seems the pandemic may make permanent changes in WFH policies even after the pandemic ends, as 31% of people predict they’ll be able to be 100% WFH afterwards, and 46% predict a partial-WFH in their future, where they’ll spend some days at home and some days at work.  On a separate friends-only post asking about the impact of WFH on productivity, I received mixed feedback from people about whether 100% WFH positively or negatively impacted productivity, and many people agreed that a partial WFH arrangement will be ideal for productivity in the future.

Interestingly, if I broke out just the people who work in tech (SW, HW, Bio, Finance, Corporate Consulting) and lived in the SF Bay Area in February, the proportions are skewed a bit more toward WFH but aren’t *that* different:

There’s been some loss of income across all sectors, including tech

It appears the pandemic has impacted the income of a significant minority of the people surveyed, about 19%.  Note that this number exceeds even the peak unemployment numbers (15%) seen nationwide.

Breaking out tech from non-tech and only including people who lived in the Bay Area in February, there appears to be somewhat less impact on income for people in the tech sector, but not drastically less.  

An astonishing number of people have left SF already

The number of respondents who lived in the city of SF dropped from 91 to 63 between February and September, for a drop of 31%.  This is incredible.  People have headed toward a wide range of new locations, including Bay Area suburbs, small towns all over the US, and cities in other parts of the US.  A few have left the country.

It’s worth drilling down on where people specifically from SF went.  Note that a total of 36 people left SF, and 8 moved *to* SF, so if you’re curious why the numbers don’t appear add up, that’s why!  A significant minority of people (11 of 36) went to Oakland, Berkeley, or Alameda.  7 of 36 have moved to small towns.  7 of 36 headed for a city outside of California. Two left the country.  

A lot of people who are leaving the bay area are going to small towns

We can put another lens on this data to see who from the Bay Area has left the Bay Area entirely.  Of the 32 people who lived in the Bay Area in February and moved out of the Bay Area this year, a startlingly large portion (16 of 32) moved to small towns (including towns in Hawaii).  People who went to other cities generally left the state entirely, with 9 of 32 moving to a city outside CA.

The tech sectors are migrating as much as the non-tech sectors

I separated out Bay Area resident (as of February) tech workers and non-tech workers to see if there’s much of a difference.  While the initial distribution of people in February was different (a much larger fraction of non-tech people lived in Oakland/Berkeley/Alameda relative to SF), the migration patterns seem to be similar, with 36% of SF tech people and 35% of SF non-tech people leaving SF this year so far.

People who suffered significant income loss were more likely to leave SF than those who didn’t, but a significant fraction of both groups still left.

About half the people who lived in SF and suffered significant income loss left the city, whereas about ⅓ of the people who lived in SF and did not suffer significant income loss did so.  Interestingly, a surprisingly large number of the latter group moved to small towns, either in CA or in other parts of the US.

What does the future hold?  It depends, but it’s going to be interesting no matter what.

Gathering information on what people would do in the future was the hardest part of the survey to do well.  The US is at a point of very high uncertainty, and it’s really unclear what the next few years will look like.  There are several major dimensions along which things could vary a lot:

  • COVID: We could have an effective vaccine in a few months, or the disease could still be dramatically affecting society a year from now.
  • The economy: While the stock market is hitting record highs, we’re still seeing huge unemployment, and it’s possible we’re headed into a recession.
  • Civil unrest: We’ve recently seen large protests in major cities in the US, and it’s unclear whether these will continue.
  • Politics: Trump may or may not be re-elected, and Democrats may or may not take over the Senate.  We’ll know in about 50 days.

Examining every combination of these circumstances would create an combinatorial explosion of different options and I was concerned that asking too many questions would cause people not to complete the survey, so I picked three in particular that people have talked about -- one in which COVID is gone and things are generally going well, one where COVID is still with us and things are going OK, and one where COVID is still with us and things are terrible.  This is the exact wording I used in my survey:

  • Scenario 1: Suppose it's June 2021, Biden is president, the economy's doing at least OK, the pandemic is a distant memory, public unrest has subsided thanks to genuine progress on social justice issues, and restaurants, bars, and clubs are open again, people are having parties etc. Where would you be most excited to live? Pick up to 5.
  • Scenario 2: Suppose it's June 2021, and COVID is still the new normal, keeping a limit on retail, restaurants, gatherings, clubs, bars, etc, but the economy is otherwise doing OK and the political situation is neutral (eg Trump is re-elected, but Democrats control both the House and the Senate, or Biden is elected but Republicans control the Senate), and levels of public unrest have dropped. Where would you be most excited to live? Pick up to 5.
  • Scenario 3: Suppose it's June 2021, the US economy is in a long and deep recession, Trump is still very much in power, the US is facing continued institutional decline, there are angry confrontations between various groups of protesters and police in major cities on a daily basis, and there's still no effective vaccine or treatment for COVID. Where would you be most excited to live? Pick up to 5.

It’s true that the scenarios reflect a negative impression of Trump’s handling of the pandemic, the economy, civil unrest, and US political unity in general, and I own that.  I had one respondent write in that Trump has been great for the economy and the US in general and thus the scenarios I chose did not make sense to them.  While it’s true that the stock market has reached record highs, there are some reasons for this that have nothing to do with genuine value being created (eg companies using tax cut money on stock buybacks, post-pandemic increases in the money supply, CARES act forgivable loans etc), and this is what it’s been looking like for people on the ground:

(source)

Six months in to a very straightforward and solvable pandemic and during a supposedly good economy, unemployment is still as high as it was at the end of Bush’s second term in the middle of a massive financial meltdown.  If you mishandle a pandemic, that’s what you get.  If you don’t think the pandemic is solvable, just ask our friends over in Asia or New Zealand for help.

In any case, back to the results.  Here’s where people want to live under the various scenarios:

While it’s impossible to disentangle all the different nuances of the scenarios, we can see some trends:

  • The pandemic has caused people to significantly re-evaluate where they want to live in the future, no matter what happens.  People are strongly considering places outside the Bay Area under all scenarios.
  • The appeal of SF, the Peninsula, Oakland / Berkeley / Alameda, and other west coast and east coast cities and suburbs is *much higher* if COVID goes away and if the economy stays in good shape.
  • Hawaii is suddenly an appealing place to live, no matter what the scenario.  This may reflect a sense of permanent shift toward WFH.  Watch out Hawaii, you’re about to get an influx of stressed Bay Areans poking their laptops at the picnic tables at your favorite beachside Kalua pork and poké stand.  3 of the 32 surveyed people who left the Bay Area this year *already* went to Hawaii, so the threat is real.  🌴😰🌴
  • Small towns in general are far more appealing than they were before the pandemic, and they’re especially appealing if COVID is still raging, for obvious reasons.
  • People are fairly interested in leaving the US in general.  They’re even more interested if the US does not get COVID under control or the political, social, and economic situations continue to deteriorate.  Of course, plenty of people make idle threats about leaving the US every four years if their preferred candidate loses the election, but in this case, 4 of the 32 people who have left the SF Bay Area already went to destinations outside the US, so I think it’s a credible threat this time.  

I also created to look at the preferences of tech vs non-tech workers on these three scenarios, and they’re relatively similar to one another.  

Some conclusions: We’re in a time of unusual flux

Social networks are extremely sticky.  It’s very hard to convince a group of friends to relocate en masse to a different city, let alone a small town in the middle of nowhere.  However, the pandemic has acted as a forcing function, pushing a lot of people to relocate at least temporarily as well as rethink a number of other things about their lives.  Separately, it’s blocked most of the common outlets that existing social groups would use to meet.  Rivers can be mighty and seemingly immobile, but when they are dammed, they find new paths.  This is what’s happening to our social networks.  Now that they’re split off into different pieces in different places, there’s much less inertia around them staying where they are now, and some re-formation is likely to occur.  This may be less true for people who are ambitious in a particular industry -- once the pandemic is over, it’s likely that it will still be easier to fundraise for your startup if you live in the Bay Area than if you’re doing pitches and product demos over Zoom -- but even that advantage may be dampened somewhat over time.  That said, if you want to stay in the Bay Area, there are plenty of people who are staying, and you’ll be in good company.  

I should emphasize that this was a relatively informal survey that’s most relevant to my social group.  There’s a much more rigorous survey process and set of analysis methodologies that I could have used, and I sacrificed accuracy in favor of speed.  However, I believe that large trends will still shine through even if the methods are not perfect.  I hope it’s been useful and I thank you for reading this.


Appendix: Demographics of respondents: