Disaggregating the economy: California

Carson Bruno wrote,

between 2009 and 2014, the Silicon Valley metro areas – a region that accounts for just 1/5th of the state’s population – accounted for 50% of California’s private industry real GDP growth.

Steve Baldwin adds,

a far more accurate assessment of [California’s] economy, [Richard] Rider writes, would be per capita GDP as compared to the rest of the country. After adjusting the GDP figures to account for the cost of living (COL), the Golden State ends up with a paltry 37th place ranking within the U.S.A., with a $45,696 per capital GDP. Even rustbelt states, such as Michigan and Ohio, have a higher adjusted per capita GDP.

There are parts of California where adjusting for the cost of living makes incomes lower than they might otherwise appear to be. Other parts of California have low incomes, but this is alleviated slightly by a lower cost of living.

It is difficult to think of California as a homogeneous economy with a single GDP factory. It is even more problematic to try to look at the entire United States through that lens.

This entry was posted in disaggregating the economy, PSST and Macro. Bookmark the permalink.

27 Responses to Disaggregating the economy: California

  1. asdf says:

    The future of California looks a lot like one of those countries with a very high performing minority/key institutions but mostly mediocrity and failure outside of that. So most places in CA will look like Mexico, but San Fransisco/Silicon Valley will be like a mini city state within it. When you add White + Asian together San Fran is less “diverse” in 2017 vs 1970. And with 44% of the population holding a BA or higher its no doubt the creme of the crop from all over the country.

    Sadly, all that talent mostly just bids up fixed quantity of real estate. I’ve heard people in SV complain that even with very high salaries their quality of life isn’t much improved.

    • collin says:

      So most places in CA will look like Mexico.

      Really? That are a lot nice places to live around California that are not third nations. And some desert/agricultural areas are very poor. Additionally, the strange aspect of the Latino growth is they are improving a lot of the old gang neighborhoods like Compton. (which was everything NWA said it was in the 1980s)

      Should smug liberals should spend more time saying what an awful place West Virginia is. Or why would any minority want to live in Alabama because they are all like Judge Moore? (Yes some do but it is not a good approach)

      • Candide III says:

        > That are a lot nice places to live around California that are not third nations. And some desert/agricultural areas are very poor.

        And that’s how you get “ruin voters” who move to Colorado and vote for the same policies that made California into the state they’d moved away from.

  2. Handle says:

    One pet peeve of mine – that appears in a lot of libertarian writing advocating for liberalizing building regulkation – is when the claim is made that poor people need to be able to live close to the city center because that’s where they can be “more productive”, but basing that assertion on some index of local nominal pay.

    This is an extremely confused way of thinking about productivity, and particularly ironic coming from people who simultaneously argue against the minimum wage (or unions lowering supply) by correctly pointing out that the higher wagers for the winners that keep their jobs obscure the losses of the people who lose their jobs.

    A barista in Mexico City earns 50 cents an hour and makes 100 coffees a day.
    A barista in Reno earns $8.25 an hour and makes 100 coffees a day.
    In one month from now, a barista in San Fransco will earn $15 an hour and makes 100 coffees a day.

    Is the latter barista “more productive” than the former baristas? Does the answer to this question change whether the wages are set by regulation or market equilibrium?

    It’s probably easier to think about in term of identical robots, maybe stamping machines, but in a world in which, for some reason, transport is prohibitively costly. Moving a stamping machine around doesn’t change the productivity of that robot. The fact that you might have to pay the robot more in one place versus another doesn’t necessarily tell you about the increased local value of the output, but it could just reflect the increased local cost of maintaining the machine. So, you could imagine the robot to be an independent contractor tasked with stamping 100 sheets per day. But if local electricity is five times more expensive, that robot contractor will demand a lot more money to stamp those sheets. That will mean the total number of stamped-sheets produced will decline to the price it takes to sustain robots using expensive electricity, and that fewer robots are built and purchased. But it doesn’t mean that any of those robots in the expensive-electricity location are “more productive” than the identical robots in a cheap-electricity location.

    Now, back to the barista example, clearly there is no increase in real productivity in terms of “coffees per day”, so the notion of increased productivity requires an analysis on the margin. But by that logic, you can suddenly make all kind of workers more ‘productive’ by imposing a quota on their industry at half the level of current production, and unionizing the workers to capture all the ‘gains’, so that the nominal price of the good or service goes way up, a bunch of workers get fired, but the ones that remain get bigger paychecks.

    Instead, when we see that one location has higher local wages than another for the same skills and real productivity, and when it’s due to market equilibrium instead of direct regulation of wages, we can probably conclude that the local cost of living is so high that one cannot recruit labor below a much higher threshold wage which compensates for his higher local expenses. That is going to make the cost of output more expensive, which which lower local product demanded and output supplied.

    Now, let’s assume that, suddenly, the local cost of living plunges because of liberalization of building regulation. Then in the absence of wage regulation we could expect average barista wages to fall as coffee output increases, and under the confused conception of productivity, we would say that the productivity of all the former baristas earning higher wages has gone way down, even though they are literally doing the exact same amount of stuff per day in the exact same spots. Which, again, is true, but only nominally amd on the margin, but that’s a very misleading way to think about the issue.

    • Tom DeMeo says:

      The answer to your question is that the barista in San Francisco is more productive. If you don’t believe that the value of something is formed by its demand at the time and place of its measure, you don’t believe that markets or economics generate any information on the nature of value.

      • Yancey Ward says:

        You just proved his point for him.

        • Tom DeMeo says:

          Not really.

          “Moving a stamping machine around doesn’t change the productivity of that robot. ”

          That isn’t true. The productivity of the robot is changed by the time and place of the demand it is fulfilling.

          • Slocum says:

            But then you’re in the absurd position of arguing that San Francisco’s development restrictions *raise* the productivity of all workers. Development restrictions -> high housing costs -> need to pay workers more to live and work there -> high productivity! It’s a view of productivity that the underpants gnomes would appreciate.

          • Tom DeMeo says:


            “But then you’re in the absurd position of arguing that San Francisco’s development restrictions *raise* the productivity of all workers. ”

            I’m not going to get into that particular argument, but theoretically, a bad regulation can raise the productivity of a service in a particular time and place while diminishing the productivity across a larger area.

            Productivity is not output. It isn’t fair. Often pure luck is involved. It is modified by good, bad, stupid and smart policy decisions. Get over it.

          • Handle says:

            @Tom: Again, I think I’ve been quite clear to distinguish between “real productivity” (as in the ratio of output produced to factor inputs in a production functions, e.g., “10 coffees per man-hour per espresso machine”) and nominal market value on the margins of a particular equilibrium. I am making an argument amount what kind of rhetoric is misleading, not an argument about the technical meaning of terms as treated by a particular discipline.

            I disagree with the rhetorical use of the phrase, “poor people moving to places where they can be more productive”, because I believe it is used with intent to mislead most of the audience. A typical person hearing “more productive” does not equate this phrase with “move to place with a higher cost of living but where they can get paid more for doing the exact same thing,”, they think of the peasant moving from a capital-poor location where he produces 10 bushels per day with a shovel to a capital-rich location where he produces 100 bushels per day with a tractor.

            Very simply, the ordinary person does not hear ‘productivity’ and ‘wages’ as equivalent concepts, even though they may be treated that was when used as economic terms of art.

            Other absurd examples of using this rhetoric seriously would be to say to a CEO thinking of moving his company to a lower cost-of-living area so he can pay lower wages, “But if you do that and move the workers to Kentucky, they’ll be less productive there.” The meaning is clearly conveyed in a better way by saying “They’ll produce the same amount of stuff, but they’ll accept lower wages for doing so.” Or if he is considering outsourcing to China, saying, “Those workers are much less productive than the equivalent American worker.”

            Or how about identical government bureaucrats, except those working in DC get a ‘locality pay’ boost. Would an ordinary person agree that they are clearly “more productive” because they are getting that pay boost, even though they are moving the same amount of paper per hour?

            Or what if we just perform the ultimate form of price discrimination and eliminate all the low-consumer-surplus consumers from the market. Voila, workers producing output are suddenly “more productive”.

          • Roger Sweeny says:

            But then you’re in the absurd position of arguing that San Francisco’s development restrictions *raise* the productivity of all workers. Development restrictions -> high housing costs -> need to pay workers more to live and work there -> high productivity!

            As Handle agrees, they may indeed be more productive in the economist sense of “marginal revenue product.” Yet the total economy will be less productive.

            Just like life on earth can cause entropy to decrease, but only because energy is coming from the sun and entropy is increasing in the total earth/sun system. (No, life does not violate the 2nd law of thermodynamics.)

          • Tom DeMeo says:


            I don’t really understand where you are going with this. I don’t know what “real productivity”and “nominal market value on the margins of a particular equilibrium” actually explains. How supply meeting demand affects productivity isn’t some kind of arcane semantic argument. It isn’t misleading. Its how things work. The productivity difference between serving a cup of coffee to a Mexican peasant and serving a cup of coffee to Mark Zuckerberg is quite real.

            In the end, I happen to think your primary point is right. It is misleading to assert that we can solve a substantial amount of poverty by simply moving the poor to more productive areas. I’m just not buying any of your analysis on productivity and confusion about what that means.

            The only reason that the libertarian writing you are citing is wrong isn’t due to confusion about the meaning of productivity. You seem to be implying that their ideas won’t work because productivity is less about circumstances than it appears to be. The opposite is true.

            They are proposing that they can move a natural equilibrium a huge amount simply by striking some housing regs when they really can’t.

            The advantage of matching a few more baristas to SF may very well be there. This is because this happens within the existing equilibrium. Time and place are powerful effects at the margin. You just can’t expect that advantage to scale endlessly.

          • Octavian says:


            “The productivity difference between serving a cup of coffee to a Mexican peasant and serving a cup of coffee to Mark Zuckerberg is quite real.”

            Not really. If the Mexican peasant moves to SF, his labor is suddenly worth a lot more; and if Zuckerberg moves to Tijuana, the value of his labor declines.

            The time and place doesn’t boost productivity: it boosts prices. You get the same service, and the same utility, for twice the price because the employee has to get paid twice as much because rent (and other living costs) are twice as high, but you’re ok with it because you get paid twice as much for the same reason.

            In other words, the price difference is basically inflation, only instead of being caused by printing money, it’s caused by the government artificially inflating the price of everything via regulation. While there is doubtless some location preference factored in, in large part (likely for the most part) the price difference has nothing to do with productivity: it’s because the purchasing power of a dollar in SF is simply less than the purchasing power of a dollar in Houston. You may as well be arguing that Canadians are more productive because they make more Canadian dollars as we do American dollars.

    • MikeP says:

      Is the latter barista “more productive” than the former baristas?

      Of course he is, provided the $15 per hour is the market wage rather than legislated wage. The people he is selling to are willing to pay a higher price for his service because they prefer to be where his service is more expensive than to be somewhere else.

      One doesn’t come to this forum expecting to see the labor theory of value so blatantly expressed.

      • Handle says:

        The difference between “nominal market value” and “real labor productivity” is precisely my point. Labor theory of value has nothing to do with it. It’s a rhetorical trick and linguistic abuse of the economic concept to say moving identical robots around to places where a marginal stamped-sheet has a higher local nominal price makes those robots “more productive”.

        If you buy this trick, then economically absurd statements that nevertheless have positive rhetorical valence such as “On average, raising the minimum wage makes workers more productive,” are just as valid.

        • MikeP says:

          Such a statement is not valid, as stipulated in my claim.

          The barista in San Francisco is paid significantly more than the barista in Reno because (a) people more want to live in San Francisco than Reno and (b) it is illegal to provide much more housing in San Francisco.

          If you have a problem with supply and demand determining the price for labor or the return to capital, then consider the barista in the lunar colony who is paid $160 an hour because he needs to buy his own oxygen and water. He’d better be producing that much value as he makes his 100 coffees a day — perhaps as a luxury consumption good for the richest colonists — or he will be replaced by a machine.

          • Octavian says:

            This is assuming the price difference is entirely determined by location preference. But artificially restricting the supply of goods and services drives up prices much as increasing the supply of money given the same level of production does. Attributing higher wages in one region to higher productivity seems analogous to automatically attributing a price increase in one country (relative to another country whose prices stay the same) to increasing relative productivity even though the former country’s money supply is expanding faster.

            In short, and artificial scarcity is similar to inflation; it boosts ‘nominal productivity’ just like inflation does, but not ‘real productivity.’ Even though SF is on the same currency as other areas, that doesn’t mean money is worth the same in SF as it is elsewhere.

      • asdf says:

        The Barista isn’t more productive. The land on which he works is more productive. That’s why most of the productivity of the coffee shop goes straight into the pockets of the landowner via sky high rents. The Barista doesn’t make more in real terms in SF vs Reno (his higher wage is also offset by his rent), meaning that they aren’t the productive component of the coffee shop.

        Addendum: It’s possible Baristas in SF might be more “productive” in the sense they are more attractive or agreeable. The high end coffee shops around me all have baristas that are white college grads who couldn’t get real jobs after graduation. They are prettier to look at and easier to talk to then if I got my coffee just a few blocks away in the ghetto. I bet they have less absenteeism and more conscientiousness as well.

        However, this often means that their rent is being subsided by their parents.

  3. Mark says:

    If someone can attain a higher standard of living by moving to Ohio or Michigan from Silicon Valley, why haven’t they done so? Baldwin refers to it as a “higher adjusted per capita GDP,” and if you’re just looking at how far your dollar goes in two different locations, then “higher standard of living” and “higher adjusted per capita GDP” are interchangeable terms.

    But alas, they are not necessarily the same. The quality of life is obviously higher for those living in SF than it would be for them living in Detroit, otherwise they would have revealed their preference by moving to Detroit where they can buy more home, better cars, more travel, etc., for the same amount of income.

    This same applies when people argue for cost-of-living adjustments based on geographic location. A federal judge living in San Francisco, it is argued, should be compensated more than a similar judge living in Ohio or Michigan. The higher cost of living in the former relative to the latter warrants compensating the SF judge more than the Ohio judge so they can both buy the same size home, etc..

    But the higher cost of living in SF is due to more people wanting to live in SF than in, say, Detroit. The federal judge in San Francisco is already compensated with the same salary because the better quality of life for her living in San Francisco relative to Detroit is more desirable.

    • MikeP says:

      Indeed. Anyone who doubts this can read Scott Sumner’s Thanksgiving piece.

      Just as candlemakers can’t comprehend the unconsidered GDP provided by the sun, so econometricians apparently can’t comprehend the unconsidered GDP provided by coastal California’s setting and climate. California’s governments skim as much of the consumer surplus as they can, but they haven’t taken it all yet.

    • lliamander says:

      Indeed, climate is something that is difficult to quantify economically and thus easily neglected.

      I would also say that culture (aka “social climate”) is an important factor. People are leaving California; partly because of the “COL adjusted per capita GDP” and partly for culture, I think.

    • Octavian says:

      By this logic, a person who lives in a country whose per capita GDP is 1/10 that of the US is, by definition, just as wealthy as an average American, demonstrated by their preference for staying in their home country, if one ignores the difficulty of immigrating.

      Indeed, if one assumes the cost of migration itself isn’t that big, this idea that location preference argument would lean one to conclude that the standard of living is equal pretty much everywhere in the world, since if any country had an inferior standard of living, everyone there would immediately move somewhere else. A 10 year lower life expectancy is just the cost of living in one’s preferred country of Kenya, and the vast majority of Kenyans just aren’t willing to trade Kenya for the extra 10 years plus having to live in Britain.

  4. Matthew Young says:


    About 5 million Californians left between 2004 and 2013. Roughly 3.9 million people came here from other states during that period, for a net population loss of more than 1 million people.

    • collin says:

      For the life of me, I still see the main reason current citizens don’t like more home building is not some kind housing price fall or short term higher taxes…

      IT is more citizens mean more TRAFFIC. It is painful here.

  5. Eric377 says:

    I do not get the problem. Yes, California is not exactly the same everywhere but if the most important characteristic of your study is that it is phenomena in California, that is just how it is. If you do not think such a study is useful, then do a study on another population of data that you think is more useful.

  6. lliamander says:

    One thing I wish the linked articles would do is publish a list of the areas in the U.S. (and hte world) with the highest “cost-of-living adjusted per capita GDP”.

  7. Justin says:

    Given that California has the most desirable climate in the country, it’s not surprising that its cost-of-living-adjusted GDP per capita should be lower. People are willing to accept a lower standard of living to live in a more desirable area. I think Steven Landburg wrote about this in The Armchair Economist in the chapter about the “indifference principle.” His point is essentially that prices will adjust so that the average person is indifferent between the possible choices. The fact that people choose to live in different places is purely reflecting heterogeneity in preferences.

Comments are closed.