Search This Blog

Sunday, July 30, 2017

Cupertino: Work-Live-Commute

Jobs and Commuting
A look at the number of employed in Cupertino and how many commute in and out.  
Cupertino Community Hall
Cupertino: Population: 60,643 (2016 est.)
Land Area: 11.3 Sq. Mi.
Density: 5,361/sq. mi.
Examining census data (as seen in the graphics below) over the 12 year period 2002 to 2014 we find that:

  1. Very few (9%) Cupertino residents (new or old) work in the city.
  2. Cupertino has added housing, yet most new residents don't work in Cupertino.
  3. Most residents (91%) commute to nearby communities.  
  4. Most (93%) of those employed in Cupertino commute in from other cities.
  5. Over the 12 year period, more employees in Cupertino commuted from further away; 28% commute 25 miles or more to Cupertino, up from 24%.
  6. Despite more housing being built, housing costs increase, commutes worsen both in and out.
Commute Times Approaching Maximum
Commute times to Cupertino may have reached a point at which no further job growth is possible. It will become increasingly difficult to hire employees as potential employees won't apply because of the commute times.  Increasing housing density would increase housing costs making it harder for many workers to live in Cupertino, but since most residents (new or old) commute out, increased housing would have little or no impact on commute times to Cupertino.

("Increased density increases housing costs" is discussed in detail here:

(The Census Bureau tool used is here with an explanation of data sources here:  Instructions on use of the census tool are here: ).

Jobs per Resident:

The Jobs / Resident ratio indicates how equitable a city's housing and jobs situation is.  If it is high (many more jobs than residents) it might indicate the city is hogging all the jobs (and tax revenue from commercial business) and letting other cities bear the expense of providing services for housing workers.  Housing requires many more services than commercial offices.

Some cities like Saratoga and Los Altos prefer to be almost exclusively residential, but most cities want as much tax revenue as they can get from business and commercial to support the expense of providing services for residents.  Cupertino is within the range of normal for most job centers in the US as is Sunnyvale. Cupertino is starting to get an excessive number of jobs per resident, but this seems unlikely to continue very long because of increasing commute times.

Cupertino: Number of Jobs / Number of Resident Workers:
2002:  26,751 / 20,243 = 1.32 = 32% more jobs than resident workers
2014:  33,949 / 23,749 = 1.42 = 42% more jobs than resident workers

Contrast with Mountain View - Number of Jobs / Number of Resident Workers:
2002:  50,174 / 31,467 = 1.59 =   59% more jobs than resident workers
2014:  85,006 / 36,935 = 2.30 = 130% more jobs than resident workers (!)

...And Sunnyvale - Number of Jobs / Number of Resident Workers:
2002:  77,480 / 58,375 = 1.33 =   33% more jobs than resident workers
2014:  84,276 / 66,209 = 1.27 =   27% more jobs than resident workers

2002 In-Out Commutes

In 2002 Cupertino had:
20,243 workers residing in the city.  Of those...
  1,950 worked in Cupertino
             I.e., only 7% of all workers in Cupertino lived there.
18,293 workers commuted out of the city = 90% of resident workers commuted out.
24,801 workers commuted into Cupertino = 93% of those working in Cupertino came from outside.

I.e., 93% of the workers commute in while 90% of the resident population commute out. Maybe it changed 12 years later in 2014? (click image below to enlarge)

2014 In-Out Commutes

In 2014 Cupertino had:
33,949 jobs in the city - an increase of 7,198 = 27% growth in jobs over 12 years.
23,749 workers residing in the city - an increase of 3,506 = 17% increase in resident workers
  2,175 residents working in the city = 9.2% of residents worked in Cupertino (down a fraction)
            = only 6.4% of all workers iCupertino live in town (down a fraction).
21,574 workers commuted out of the city = 91% of resident workers commuted out (up 1%).
31,774 = 94% of workers commuted into Cupertino from other cities.

I.e., In 2014, 94% of the workers commute in - about same as in 2002  - while 91% of the resident population commute out.  (click image below to enlarge)

Most residents (80%) of Cupertino  Don't Work There

Over the 12 years 2002 to 2014, there was new housing for 3,500 more workers in Cupertino.  Yet only 225 more Cupertino residents worked in Cupertino.  Despite increased housing, 3,250 more residents commuted out and 7,000 more commuted in.

Almost 7,200 more jobs were added.  Despite all those new jobs the percentage of residents who commuted out stayed about the same over 12 years.  Twelve years is plenty of time for residents to find a job in their home city of Cupertino if they want to.  Building more housing doesn't do anything to lessen commuting in or out.

It appears that for the vast majority of people, living and working in the same town isn't as important as finding a job that suits them, even if it means commuting.  They may live in Cupertino because they like the schools, or it is equidistant from jobs for both husband and wife, or any of a number of other reasons.  Commuting is not enough of a factor for them to move either home or job.

Commutes to Cupertino Get Longer

What we find in the next pair of data visualizations for 2002 and 2014 as seen below is that most people are commuting to Cupertino from further away in 2014 than in 2002.  (Yellow high-lighted numbers below.)  In 2002, only 6,400 (3458 + 2,938 = 6,396) or 24% commuted more than 25 miles.  By 2013 that had risen to 9,400 (5,320 + 4,074 = 9,394) or 28% (commuting over 25 miles) into Cupertino - nearly 50% more.  Nor can that increase be fully explained by private buses to San Francisco.  Further below we see figures showing the largest group commuted from San Jose - only about 1,000 (about one third) of that increase came from San Francisco.

Summarizing census data shown in visualizations below: (click table to enlarge)

The above data is from the two charts below.  

The 2 "radar" graphs show most everyone is commuting in from the North, South-East, and East.  Cupertino seems to be on the edge of a commute area.  The other thing to notice is that the commutes have increased in distance.  The number commuting has increased in all four distance categories but the longest commutes to Cupertino have increased more.  The "bike-able" Less than 10 miles category for commuters to Cupertino decreased from 50% to 48% while the two 'More than 25 miles' categories increased from 24% to 28% of commuters.  This suggests that Cupertino employers (mainly Apple) are having to go further and further to find employees.  Employees in Silicon Valley change jobs frequently and are not going to uproot their family just so one parent can bike to work.

Commuting on a luxury bus is better than driving but at some point the sheer time involved becomes a burden.  Getting home at 7:30 is a lot different than getting home at 6:00.  The commutes become too much for employees. Companies will relocate or expand elsewhere to find workers.
See your kids awake only on weekends?

Vehicle Miles Traveled:

Vehicle Miles Traveled (VMT) is the standard way of describing traffic.  For example, if 10 people each commute 12 miles, that is 10 people times 12 miles = 120 VMT.  How many VMT are logged by commuters to Cupertino is in the following table (click to enlarge):

The high-lighted last column illustrates that the biggest percentage increases were in further out commutes.  This is probably the motivation for Apple to start setting up a big campus in north San Jose.  Commutes to Cupertino are getting increasingly time-consuming for the average worker.  Put another way, Apple has run out of nearby workers so they have to go further out.  The longer commute likely makes it harder to hire.


Biking is very healthy and enjoyable.  Commuters who walk or bike to work actually enjoy their commute.  Cities should encourage it.  Does it make a difference in pollution, congestion, or GHG emissions?
Philadelphians Bike To Work
The last line in the previous table shows what VMT becomes if everyone who is within a 10 mile radius bikes to work (10 miles is about an hour by bike).  That is about 50% of all Cupertino commuters.  Most bike-to-work advocates would be ecstatic if they could get 10% to bike to work (second to last line in table above).  The impact of getting 10% or 50% of commuters to bike to work is seen in the following chart: (click chart to enlarge)

With 10% of all commuters biking to work we get a small but noticeable reduction in VMT.   With 50% biking we would get a noticeable (about 14%) reduction in VMT, but not really a "save-the-planet" or "clear congestion" difference.

This illustrates the "go big or go home" aspect of trying to reduce GHGs, pollution, or congestion by individual actions.  To get a meaningful reduction in GHGs we need all transit (including cars) to go fully electric with all electricity generated by renewable energy with some sort of storage (battery, pumped hydro, etc.) to fill in fluctuating usage periods.
Tesla Model 3 - 100% Electric - Saves the planet
So in answer to the question a few paragraphs ago: biking, despite its many virtues, does not make much difference in congestion or GHG emissions, even in the most extreme cases.  The arguments for biking should concentrate on child safety, health, and making life more enjoyable - all valid reasons to encourage biking.

2002: Where the Jobs Are

The blue "heat maps" below show where the jobs were most densely concentrated in 2002 and twelve years later in 2014.  The actual numbers have increased in all job centers so the maps show relative concentrations of jobs.  There are no surprises.  (click on chart to enlarge).

2002 - Where the jobs are
In 2002, the jobs were mainly concentrated in Cupertino, San Jose's North First St., Palo Alto, Sunnyvale, and Santa Clara.  The data in the lower right corner of the graphic above shows two out of three (67%) Cupertino residents were able to commute less than 10 miles.

This is a much greater percentage with short commutes than is typical in the area.  It largely explains why Cupertino is so expensive.  People pay more to have shorter commutes.  The people who can pay more tend to have better educations.  This means their kids have the social and financial support to do better in school so the schools are "better" reinforcing the attraction of Cupertino to educated people in a self-reinforcing cycle.

2014: Where the Jobs Are

By 2014, Google had become a major jobs presence, Apple had grown tremendously (the iPhone came in 2007).

The more intense blue in Mountain View in the 2014 heat map below shows that Mountain View became a much more major job source relative to Sunnyvale, Cupertino, and San Jose's North First St.  Sunnyvale grew the number of jobs but not as fast as Mountain View - the decline was not in absolute numbers but relative to other nearby areas.

2014 - Where the jobs are

The data at the right of the visualization above shows that the percent of Cupertino residents who commuted nearby actually decreased from 67% to 64% while the number who commuted more than 25 miles increased from 15.7% to 18.2%.  This runs contrary to the narrative that building housing and increasing density for people allows shorter commutes.  It also runs counter to the narrative that adding jobs to a city lessens commuting distances.  Could it have increased commuting distances by Cupertino residents?  Perhaps building housing and adding jobs did nothing whatever.  Perhaps the number of jobs around the SF Bay area increased randomly with respect to geography.  Many people seeking jobs consider first, how it helps their career, and only second, commuting distance.

Where Workers Commute From:

The graphic tables below show that the distribution of where workers commute to Cupertino from is relatively constant.  Most come from San Jose, followed by Sunnyvale and Cupertino.  The difference between 2002 and 2014 is that a smaller percentage of workers are commuting to Cupertino from nearby, and more commute from further out.  The numbers from all cities have increased but disproportionately greater from more remote communities.  Sunnyvale had 400 more commuters to Cupertino than 12 years before and Mountain view added 200.  Note that more Sunnyvale residents worked in Cupertino than did Cupertino residents (2,540 vs. 2,275).  Not huge amounts, but it shows falsity of the idea that if cities build more housing fewer will have to commute.  The direct opposite occurs.  (click on graphics to enlarge)

2002                                         2014

The visualization below shows a 10-mile radius circle centered in downtown San Jose.  For 2014, the data on the lower right of the chart shows a jobs / worker ratio of 1.05 (703K / 669K) - very close to one job per worker.

10 mile Radius from Downtown San Jose
Extending the radius of the circle from 10 miles to 15 miles (below) includes all of Mountain View and parts of Palo Alto and Fremont.  

15 mile Radius from Downtown San Jose

The number of workers goes up nearly 100,000 but the number of jobs goes up nearly 200,000.  This shows the extreme imbalance of the jobs vs. workers in the area between Santa Clara and Palo Alto.  Now the jobs-to-worker ratio is 1.17 which is still low or moderate.

There is much discussion about teachers not being able to buy a house in the towns they teach in.  The solution is not to build more housing - that only increases the demand for land thereby increasing the value of land thus making housing even more expensive.

If increasing density by building "up" lowered housing costs, why does a Manhattan studio condo with no parking cost $615,000...

... the same as a 4 bedroom house in the suburbs?

The solution is a more logical and also equitable distribution of jobs.

The peninsula area from Santa Clara to San Mateo will in the foreseeable future be a center for start-ups.  That area is close to the venture capitalists on Sand Hill Rd. in Palo Alto who don't want to go very far to check up on the start-ups they fund.

(From Wikipedia:  "Sand Hill Road, often shortened to just "Sand Hill",[1] is an arterial road in Menlo Park, California, notable for its concentration of venture capital companies.[2] The road has become a metonym for that industry; nearly every top Silicon Valley company has been the beneficiary of early funding from firms on Sand Hill Road.[3]" )

It isn't a matter of how high or densely you can build but how do you move people in and out?  You could, in theory, house all Apple workers right next to work (see image below).  And what would Wolf Road and Highway 280 look like when they decide to go shopping or go to a local park?  (Click image to enlarge)

“Of course, the bigger question,” says Jones, “is why the employees would want to live in a community like that.”  from:
Cupertino is on the periphery of the populated area.  It is bounded by protected open space on one side and equally densely-packed small cities on the other.  There is a point at which further job growth is not possible without extreme strain on housing and transit - which we are currently seeing.

The following chart of a 10 mile radius around Cupertino shows that for the area as a whole, (Mountain View, Sunnyvale, Santa Clara and parts of San Jose and Palo Alto), there are many more jobs than workers.  The ratio is 1.5. That is because of the large spaces to the south west of Cupertino with very few people.  Moving the center of the circle to Central San Jose as shown above, largely takes care of that problem.

Just looking at the the numbers living inside each 10-mile radius circle, we see 669,000 living in the circle centered around San Jose, but only 487,000 living in the Cupertino 10 mile radius circle.  This is a 37% increase in the number of available employees.  More importantly for the employees, it makes it easier to commute from more affordable areas.  This by itself should make central San Jose more attractive to employers.  Sure, there are plenty of engineers in the Cupertino and surrounding area, but you also need accountants, clerks, security guards, maintenance workers, etc., etc.  Being within a desirable commute area is good for everyone.

Building 'up' is not a solution to housing price inflation, as seen by looking at the densest parts of major cities like London and New York.  The desirability of living in the center results in the bidding up of land values with concomitant higher rents and prices.  It is no accident that a nice single family townhouse in Mayfair in the center of London costs $7M.  Land is the limiting factor and we can't make more of it.

Lower Manhattan
Cheap Housing?

No matter what some may say about changing attitudes, most people do not want to live in a condo high-rise.  They want a single family home and will commute as far as necessary to get that.  We see that already with 93% of new workers in Sunnyvale commuting from Morgan Hill, Livermore, Fremont, and Pleasant Hill. (c.f.,


The map below shows the urbanized areas of the San Francisco Bay Area.  Cupertino is clearly on the very edge.  Which implies it is on the edge of the easy commute area.  The area to the west of Cupertino is simply not going to be urbanized.  As we have seen, it is getting increasingly difficult to commute to Cupertino so Apple has to expand elsewhere in order to continue to grow.  Google is in the same situation.
Cupertino is on the very edge of the Urbanized Area
Based on census data for the last 14 years,
  1. If Cupertino adds housing, much of the new housing will go to people who commute out of Cupertino, giving no relief to those commuting into Cupertino.
  2. Increasing housing density would further increase the cost of land and therefore the cost of housing built on the land.   ("Increased density increases housing costs" is discussed in detail here: and here )
  3. Cupertino is already very expensive so further increases in housing costs would be unmanageable for most of those who rent there.  They would have to leave to be replaced by more affluent workers who can afford the higher rents.  This would cause even longer commutes by the less affluent from lower rent locations resulting in greater difficulty in hiring people to work in Cupertino.
It appears Apple as the main employer in Cupertino has no choice but to expand closer to where workers are.  They will be doing just that with a new campus in San Jose north of the airport.  See below:

86 Acre Future Apple Campus in San Jose
As Apple and Google expand in San Jose, housing costs in Cupertino and other communities will stabilize.  Housing prices won't decrease because most residents of Cupertino don't work there.  People will continue to pay high prices to be in the highly rated Cupertino school district while commuting to other cities.
What Home Buyers Really Care About
As downtown San Jose develops, housing prices within central San Jose will increase because people pay more to be close to major job centers.  San Jose has 180 sq. miles and about 1,000,000 people.  Cupertino has 11 sq. miles and 61,000 people = 6% of San Jose's area and population.  San Jose can absorb the influx much more easily than Cupertino.  A few neighborhoods near downtown and transit depots will see an increase in property values, but most neighborhoods won't even notice.  Schools near the new high tech job centers will 'improve' not because of anything teachers do but because the students will be the children of more highly educated parents.

Over time - and I mean years - more and more engineers will find housing near the new job center in central San Jose to be closer to work.  More and more companies will locate there to be near where they have access to more engineers.  Thus starts a self-reinforcing cycle making San Jose a big tech center.  Eventually, as office building leases expire and older employees retire, it will be easier for Apple and Google to relocate most of their headquarters functions there as well.

San Jose will evolve into a major city with a vibrant and exciting cultural life that people from the boring suburbs of Cupertino, Sunnyvale, and Fremont can visit for an evening or weekend before returning to their quiet little homes.