When I got my first summer job, my coworkers and I were paid the same hourly wage. Doing sales and booking in three languages for a shipping company, we spent our days mostly managing fares and talking to calling customers over the phone (this job has since been all but automated by well-functioning e-commerce solutions, since customers valued the trip rather than our hyper-friendly voices).
The wage structure was transparent and rigid (and the numbers somewhat outdated): $12.75 an hour, another 25 percent for evening shifts, and double pay on Sundays and public holidays.
Naturally, getting the Sunday, holiday, and night-time shifts was all we quibbled over, seeing as how that was an easy way to increase our take-home pay. Even long-term employees with decades of experience were on these pay scales, with some extras for leadership positions and different responsibilities as well as union-negotiated perks for long service (say extra days off and fringe benefits for employees over a certain age).
I always wondered why that was. I was fresh out of high school; one of my coworkers was an impressively seasoned traveler; another spoke fluent German and had five years of experience in the retail traveling sector. I merely spoke decent German and was good at navigating the booking system faster than my somewhat older colleagues. Surely, I wondered, their labor was worth more than mine, as they could deliver higher-quality service than I could?
If we read Izabella Kaminska’s article in the Financial Times last week, we might get the impression that my former employer made a huge mistake — not for failing to match our pay to productivity, but by not properly smoothing our incomes over various phases of our lives. In contrast, Kaminska invokes a rosy idea about a long-lost past where staying with the same firm for one’s entire career allegedly made for a more harmonious labor market — an implicit social contract between employers and employees. In her account, workers go through phases of their lives with fluctuating productivity that companies used to smooth over and thereby effectively stabilize take-home pay over the course of one’s career; in my low-productivity years, I was “overcompensated,” paid for by “undercompensating” me during my high-productivity years.
Beyond merely voicing this idea, Kaminska uses it to condemn the gig economy; the “hyper-flexible model” of Uber and others allows companies to cherry-pick their most productive workers and abandon that rosy model that Kaminska imagines held true in the past. It’s unfair, she claims, as
in reality, nobody other than a Marvel comic hero is able to maintain the same level of productivity over their entire lifetime…. Gig economy supporters might argue that workers should set aside income to account for their low-productivity phases themselves. But this is unfair. It will always be harder for an individual employee to manage productivity-related income inconsistency than for a corporation that employs multiple people in differing stages of life. The capacity to smooth inconsistencies is a function of scale and expertise.
This section should strike the reader as somewhat odd. First, it’s not at all clear that we have fluctuating productivity over time — particularly not in the fields affected by the gig economy. Perhaps the labor supply curve shifts over time but in this case not from worker productivity. A lawyer, an accountant, a consultant, or other office-bound and relationship-based middle-class professions might see the kinds of productivity fluctuations over their careers that Kaminska is concerned with — working “unpaid extra hours to get ahead” or “subsidise our employers by using our own equipment” — but that hardly applies to the incumbent delivery workers or cab drivers that the gig economy is directly challenging. In most countries, cab drivers protected by regulation are paid by shifts (essentially day rates), regardless of how many trips they make or how well they serve their customers. The productivity of those workers is unaffected by varying life phases, at least in the kinds of examples Kaminska discusses.
Second, her reasoning that corporations’ had the ability to smooth changes in productivity over workers’ lifetime only makes sense in the context of rigid and heavily regulated labor markets of the 1950s where employees rarely left for a job elsewhere. The very meaning of smoothing involves overpaying during low-productivity phases and underpaying during high-productivity phases; as firms now know that employees are unlikely to devote their full career to that firm, the firm can no longer take advantage of that smoothing strategy.
Training employees in non-firm-specific skills suffers a similar drawback: it makes little sense to invest in the human capital of your employee unless you can ensure that the future productivity value of those skills accrues to you. Perhaps that’s a deadweight loss, but it’s induced by workers’ career goals rather than firms’ efficiency scrambles.
Third and more crucially, “managing productivity-related income inconsistency” requires the ability to minutely monitor productivity. Until not long ago, meticulously measuring an individual employee’s contribution to a project or a firm’s bottom line was either impossible or prohibitively expensive. Firms elegantly solved this by averaging among employees, accepting the resulting inefficiencies because the utopian scenario of precisely segmenting pay was not available.
Like the shipping company I worked for, it simply wasn’t worth the effort to figure out which of us could handle more clients in a day, how well we did so, or how often we sold additional services and segment our pay scales accordingly. The reason it wasn’t worth the effort wasn’t structural or economic, but because the cost of acquiring, analyzing, and implementing that information was too high compared to the minor differences among us.
Technology entirely changes that factor. Monitoring and quantifying productivity differences between workers was hard and expensive for my employer then, but is effortless and crucial for Uber now. Indeed, the numbers are right there, in plain sight: how many trips the driver made, how well those paid, how much he received in tips, and what rating customers gave.
Whereas the productivity information was previously expensive for the employer to acquire and difficult for the employee to credibly signal, technology-heavy companies like Uber have lowered the obstacles for both. It extends to many other industries too: profitably broadcasting radio shows used to require large fixed costs up front, spread over many different shows; today anyone with a computer and internet access can make a podcast — and, with a large- or specialized-enough clientele, generate decent ad money.
Notice how this technology push also gets at a more profound question in economics: the theory of the firm. With Uber, a driver needs much less overhead to pursue his business; a podcaster almost none at all. Adam Smith famously wrote that the division of labor was limited “by the extent of the market,” and Ronald Coase’s article “The Nature of the Firm” asked why, if price signals and markets are so efficient, we organize production within firms. This line of inquiry, earning Coase a Nobel Prize in 1991 and his follower Oliver Williamson another in 2009, resulted in what we now call transaction cost economics: market prices are efficient in allocating resources across the economy, but exchange involves some trading costs before it can take place, such as the effort and money spent in searching and matching clients and workers to companies. The role of firms is to sidestep this otherwise efficient allocation mechanism when costs of exchange are prohibitively expensive, internalizing the production in an organization — the firm — shielded from market prices.
New firms embody innovative approaches in production, goods, or formats. Additionally, new technology allows firms to do things they couldn’t do before — in this case monitor and assess the relative productivity of workers. In the gig-economy case, the barriers to trade that bound employees together with capital owners in firms have shrunk dramatically; when technology allows for economic agents to transact via market prices rather than market-shielded firms, outsourcing the subdivision of one’s production becomes both accessible and desirable.
Transaction costs have fallen, and production is being re-organized. Technology changes the way we do things, primarily by allowing us to do things that were too expensive to pursue before. That’s a wonderful and productivity-enhancing thing, not something to lament.