How Much Work is Real?

Epistemic Status: Casual, just framing things

Some work is clearly “productive.” If you plant things in a garden, you put in work, and you get out plants.  If you cook a meal, your family gets fed. If you build a building where people want to live or work, they get shelter. If you treat a patient, the patient gets better. If you carry goods to the place that they’re sold, people get their stuff. If you invent a labor-saving machine, people get to free up their time for other things.

Productive work creates value, in the sense of “doingstuffness”, mana, usefulness-to-humans, etc. It’s not just effort expended, or an accounting formalism like dollars, it’s an increase in the “real wealth” of humanity. That’s not a well-defined concept, but it’s worth pointing at, so we know what we’re complaining about when we see deviations from this.

In the standard capitalist story, you get paid for work because you created value for somebody; they wanted your stuff so much that they were willing to give you something in exchange for it.

In this world, all productive work is honorable.  Work is fair — on average, you get what it’s worth — and it’s a contribution, however small, to the wellbeing of humankind, the fire that beats back against the blackness of hard vacuum.

But there are ways that things called “work” can fail to be productive work.

Fraud or crime obviously are not productive. If you get money from people by tricking or terrifying them, you’re not getting it by providing them with value. You’re not a maker.

Enforced monopoly power is also not entirely productive. If people are required by law — on pain of punishment — to buy your product, then at least some of your revenue is driven by fear, not desire.

Regulations can be a form of enforced monopoly power. If only people who meet certain criteria are allowed to sell, then people are buying from you and not your competitors not because they like you better, but because your competitors are driven out by fear of punishment. Once again, you’re profiting partly off fear, not just desire.

A job that is funded by taxpayer money, or by the fact that the product sold is mandatory to buy, or by the fact that nobody knows whether it would be illegal to get rid of that job and they want to play it safe, doesn’t need to be useful at all.

And there are still more indirect ways that a job can fail to be useful.  If you sell to people who don’t do anything useful, then your job would not have been necessary in a sensibly organized world, even if you do nothing dishonest yourself and genuinely add value to your customers.

This is what it means to live in a “mixed economy.”  Not everything that everyone does for a living is genuinely useful.

If there are bullshit jobs, as anthropologist David Graeber claims, then that’s a shame, from the perspective of human well-being. If we have enough real wealth, enough mana, to support even people who aren’t making mana, then why not just allow leisure, instead of forcing people to go through the motions of dull and unnecessary work?

This is largely the position of left-libertarians like the people at Center for a Stateless Society.  They make the empirical claim that most of the present economy in developed countries is coercive and unproductive, the result of crony capitalism and regulatory capture rather than honest, useful work.  As such, a “freed market” without such corruption would actually be more egalitarian than our current economy.   Since government promotes monopoly, Big Business wouldn’t be sustainable without coercion.  Since highly regulated and positional goods like housing and education are essentially mandatory for participating in much of modern life, if those mandates were abolished, socioeconomic inequality would drop.

On the other hand, this empirical claim could be false. Nobody denies that some corruption exists, but it might be the exception rather than the rule. We might not, in fact, be in post-scarcity conditions.  So-called “bullshit” jobs may actually be valuable, just easy to dismiss by outsiders like Graeber.  Growing wealth inequality may be largely the result of winner-take-all phenomena, as Tyler Cowen thinks — in his model, the working rich really are more productive than ever, thanks to the amplifying effects of technology.  Love it or hate it, says Cowen, capitalism works the way it says on the tin.

There is some evidence that economic rents are on the rise in the US. Wealth inequality has risen over recent decades, but labor productivity has declined. Economic dynamism — the number of people changing jobs and starting new businesses — has also declined.

One important piece of evidence that rents are on the rise in the United States is the divergence of rising returns to capital and declining real interest rates. In the absence of economic rents, the return on corporate capital should generally follow the path of interest rates, which reflect the prevailing return to capital in the economy. But over the past three decades, the return to productive capital generally has risen, despite the large decline in yields on government bonds.

Other firm-side evidence points to an increased prevalence of supranormal returns over time. Between 1997 and 2012, market concentration increased in 12 out of 13 major industries for which data are available, and a range of micro-level studies of sectors including air travel, telecommunications, banking, and food-processing have all produced evidence of greater concentration.

The fact that variations in the rate of return to capital have increased enormously across firms may also at least partially reflect increased concentration and the role of economic rents. Finally, there is evidence that land-use regulation may also play a role in the presence of increased economic rents, decreasing housing affordability, and reducing nationwide productivity and growth by restricting supply.

There’s also Matt Rognlie’s paper showing that the long-term rise of capital’s share of wealth (compared to labor’s) is almost entirely a result of increased housing prices — literal rents, kept high by land-use restrictions.

And there’s the phenomenon that S&P 500 firms are now 5/6ths “dark matter” — that is, things that a new entrant to the field can’t copy.

Imagine that you wanted to create a new firm to compete with one of these big established firms. So you wanted to duplicate that firm’s products, employees, buildings, machines, land, trucks, etc. You’d hire away some key employees and copy their business process, at least as much as you could see and were legally allowed to copy.

Forty years ago the cost to copy such a firm was about 5/6 of the total stock price of that firm. So 1/6 of that stock price represented the value of things you couldn’t easily copy, like patents, customer goodwill, employee goodwill, regulator favoritism, and hard to see features of company methods and culture. Today it costs only 1/6 of the stock price to copy all a firm’s visible items and features that you can legally copy. So today the other 5/6 of the stock price represents the value of all those things you can’t copy.

“Intangibles” would certainly include rent-seeking forms of favoritism.

It also includes patents, which arguably do not increase innovation on the margin (according to natural experiments between different countries with different patent regimes or changes in patent laws.) Copyright and patent lengths have gotten longer in the US, and patent applications have grown at an accelerating rate; the growth of intellectual property is another example of our economy becoming more monopolistic.

How much of our economy consists of rent-seeking would be hard to detect, and I’m not sure anybody has attempted it. “Spikiness” in wealth between firms or individuals could be either due to monopolistic privileges or variance in productivity.  Concentration of growth in highly regulated industries points towards rent-seeking being prominent, especially if the measured outcomes of those industries don’t improve (e.g. healthcare), but it doesn’t tell us what percent of the value of healthcare is due to rents.

And, of course, even such estimates don’t tell us what to do, because of path-dependency effects. Even if we discovered with high confidence that an industry was mostly corrupt, that doesn’t guarantee that “anti-corruption” efforts will actually make it less so.  (Sometimes the increased administrative demands of making sure nobody gives bribes cost more than the bribes themselves did.)

But the question is relevant, to those of us who want to know “where can I find productive work?” and “how much misdirection is going on under the surface of today’s world?”

I work at a biotech company. I made a special effort to find a job that was as honest as possible, while still being in my field.  And I think we are honest here; the official purpose of the company (to find new promising drugs) is also the implicitly endorsed goal that people actually work towards.  We’re a bunch of scientists, with scientific sensibilities. But we’re still in an industry defined by grants, patents, regulations, and other monopolistic practices.  There are still, I think, pockets of inefficiency that result from being in that industry.  Bigger, older businesses often get flummoxed when their startup partners move too fast.  And those of us who don’t work in the lab don’t really need to work 8 hours a day every day in order to meet our planned goals.  My job isn’t bullshit, by any means, but I sometimes suspect that it isn’t maximally productive.

I don’t believe in being so obsessed with personal purity that you never get anything done — that’s not useful and it’s not the point. It’s more about trying to figure out what kind of world you live in.


22 thoughts on “How Much Work is Real?

  1. This is what it means to live in a “mixed economy.” Not everything that everyone does for a living is genuinely useful.

    If there are bullshit jobs, as anthropologist David Graeber claims, then that’s a shame, from the perspective of human well-being.

    I think this is missing something important. There can be market work that is useless – famously, half of advertising (but we don’t know which half). There can be nonmarket work that’s useful, because of market failures or simply because something that could have been provided in a market context is provided by the government or a nonprofit or a regulated monopoly instead.

    Likewise, whether your gains are ill-gotten isn’t necessarily tied to whether your work is productive. A government-enforced monopoly can sometimes be reaping illegitimate profits, but this only works because people need the product. The harm is preventing others from doing the work – the work itself is good and diligently increasing the quantity or quality of a monopoly’s output can be good.

    This feels similar to something I think Graeber gets wrong about bullshit jobs. He focuses on the extent to which jobs *feel* like bullshit, but that’s not the only way in which we might be able to take it much easier than current practice. There’s also a dynamic where legitimate work is stretched over many more people than are required to do it. For instance, in the era of strong labor unions, restrictive work rules were often designed to spread the work over many people in order to keep more people employed. Most line workers were still doing real things, they were just idle a lot more than they had to be. There were presumably a smaller number of people enforcing this idleness, who had more obviously useless jobs.

    Overall this suggests to me that using “how much demand is there for my work?” in the form of price signals as a proxy for productivity isn’t the full answer, and the right thing to do is to additionally just look and see for yourself whether you’re doing anything interesting.

    • > “There can be market work that is useless – famously, half of advertising (but we don’t know which half).”

      Is there a reason you’re rejecting measuring value in expectation here? The portfolio as a whole is producing some known value. We know, in principle, that we can produce more with a better arrangement, but we don’t know what that arrangement is.

      > “Overall this suggests to me that using “how much demand is there for my work?” in the form of price signals as a proxy for productivity isn’t the full answer, and the right thing to do is to additionally just look and see for yourself whether you’re doing anything interesting.

      I hear things like this a lot and I worry that people grossly overestimate their ability to figure out an efficient allocation of resources. Every time I’ve seen some attempt this in person, I’ve been able to find an important factor they’ve neglected (the most common mistake is seeing the purchase of insurance-type things as a mistake when no crisis happens, or the failure to buy insurance-type things when a crisis did happen). The level where price signals (or similar mechanisms) provide significant value that all but the most talented managers miss seems to be in the low dozens of people.

      I think people should ask the question of “am I doing anything interesting,” but for reasons of life satisfaction and measuring the type of impact you’re having. I suspect that asking this question will increase your error rate in distinguishing useful and useless work relative to just looking at more objective “rent seeking” signals.

      • I agree that people often wrongly round the value of indirect or conditionally valuable work down to zero. And getting paid a market price for a thing is substantial evidence that you’re adding value. But varying levels of competition, consumer surplus, and positive externalities, and Goodhart’s Law all suggest that there are large discrepancies between share of goodness produced and share of income produced.

      • Is there a reason you’re rejecting measuring value in expectation here?

        I’d like to be able to talk about heterogeneity at all (i.e. not just infer that because work does good on net, all work is useful in expectation) and I don’t know of a principled reason to stop at the boundaries of something like the category of advertising.

      • Agreed.
        Looking at value in expectation is really important. There have been a series of posts with a strong “makers producing tangible things have real value while others do not” kind of vibe (which I am sure wasn’t intended to devalue other’s work just to express the legitimate excitement of making things). But I do think this framework misses areas where value is created that can only be understood through expectation.
        Something can be real without being tangible. If someone reduces the probability of some negative event from 2 per 100,000 to 1.5 per 100,000 and you have 300,000,000 people impacted than that’s a lot of value; but a person can’t see or feel, or even non numerically conceive of 0.5 per 100,000. And a person can’t see or feel, or even non numerically conceive of 300,000,000 (after around 50 or so the brain is just at “many”).
        A lot of activity in the modern world in areas from regulation, to insurance, to finance, to process management at large corporates involves small improvements that are applied to huge numbers. The tangibility framework is suspect in these contexts.

  2. It’s also technically possible for literally everyone to be paid less than the total value of their work (because positive externalities that don’t show up as GDP), AND for some industries to be disproportionately capturing rents.

    • Actually, that’s only true for PAID work. Home production (e.g. tending your garden, housework) is generally unpaid & doesn’t count towards things like national income or GDP, but it’s clearly getting people nice things.

  3. I’d be interested in an analysis of how much economic growth is real. Offhand it seems to me like growth is happening in healthcare, education, real estate, and finance sectors. All these have very deep ties to the state. I’m most definitely not a libertarian, and I don’t buy the idea that people freely maximizing their own well-being is a good idea – partly because I don’t buy the idea that ‘maximizing well-being’ is a good goal, partly because I think in the first place people derive well-being from group rather than individual activities, and communal rather than private goods – but I recognize that at least ‘people freely chose to purchase this good/service’ is a limited indication of its usefulness.

    People don’t ‘freely choose,’ by any stretch of the imagination, their healthcare, their education, their retirement plans, or their housing. These sectors need to mount some other defense of their sprawling size and growth. Are schools making people smarter or more knowledgeable than they did back when we were spending much less on them? Are hospitals giving people more useful lifespan? Is the finance sector providing more security and stability than it once did? Are houses really that much better than they were a few decades back?

    • BTW, do you have a blog or something? I just noticed that you’re a regular commenter and seem to have an interesting take on things.

      • Not yet – I’ve been thinking of starting one. Probably will, soon. I’ve been weighing whether it would be better to use my real name or stay anonymous.

  4. Libertarians and the like would probably claim that enforcing of government regulation is non productive or a bullshit job. But you would be hard pressed to find anything that has been as value added to society as the reductions in ambient lead levels and lead exposure engineered over the past 50 years.

    Be very wary of measures that miss externalities, frictional costs, etc. Its very easy to have a simple theory if you don’t mind being wrong.

    • Right, I’d classify lead reduction as a positive externality. This is related to Ben’s point about how actually the valuableness and the legitimacy of an activity are different. For an extreme example, Communist Russia and China *did* produce value when they industrialized — just at a cost most of us would find unacceptable. “Consensual” is a proxy for “useful” but it’s not exactly the same thing.

      • I tend to think of “consensual” as a safety constraint on behavior, and “useful” as an overarching goal. This may not be the most elegant setup but it seems to have some correlation with sanity?

  5. The analysis in The Happy City is slightly naive/myopic, but I think talking about something real and very very large:

    I think of frames like that almost the same way as Marie Kondo. Ontology problems on a societal level. In your room it results in piles of uncategorized items and shows up in your life as slight friction to lots of things and a general background anxiety. In a society it results in weird legislation and shows up civilizationally as no one knowing what to do about certain problems.

    • I think you’re talking about a thing I think of as “buffer zone inflation.” When you’re traveling, say, as an affluent adult, you go along well-worn and rather wasteful paths, you throw money at problems and don’t experiment, so that unknown-unknowns don’t bite you. If you’re hiking off-trail, or traveling cheap, you can see exactly where the constraints are, and it’s not actually that dangerous to veer up near them. My natural disposition as a younger person was to do low-margin-for-error stuff; as I get older, I find myself leaving more buffers. Which is more inefficient, and makes me feel padded and boring. “Leave extra time/money/space for unforeseeable bloat and catastrophe” is locally sensible but isn’t a thing I feel good about.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s